MG-Cars.com
MG Midget and Sprite Technical Rev counter BBS discussion at MG-Cars.net
MG-Cars.net
Welcome to our resource for MG Car Information.
Recommendations
Parts
MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.
|
|
MG Midget and Sprite Technical - Rev counter
| Hi Everyone I have changed to electronic ignition and now the rev counter is not registering the correct revs the needle moves but not enough I presume I need to mess with the electrics to rectify this does anyone have a fix |
| R Deans |
| yes- but to save confusion , are you on negative earth |
| William Revit |
| And have you got an RVI or RVC rev counter. My guess is you have an RVI. |
| anamnesis |
| tacho doesn't matter what brand ignition --part no. |
| William Revit |
| Have a look at the website of Clocks4Classics they may have a solution. |
| Roger D |
| There's an easy solution but we need to know what brand ign. RD has and if it's pos or neg earth first. |
| William Revit |
| RVI rev counters generally don't work with eletronic ignition. Yep, useful to know pos or neg earth. That's why I said "AND" have you got an RVI or RVC rev counter. And being a Spridget, pound to a penny, it's a Smiths. |
| anamnesis |
| Ah, misread Willy. Yep what type and brand ignition. 😁 |
| anamnesis |
| Ok I will check it all out tomorrow and get back with details |
| R Deans |
| RD it's a simple fix- but no use spitting all sorts of stuff out here until we know what you have there and what year model is your car an--you're barking up the wrong tree and confusing things with the type of tacho--it's not an issue, it's how it's wired that matters. |
| William Revit |
| As you say Willy, until it's known what he has, the cause and solution isn't clear. I haven't proposed a solution yet. I've asked a question and made a comment; backed by personal experience, and the experience of others. How it's wired may or may not fix it. It depends on the type of electronic ignition; personal experience. |
| anamnesis |
| I have an Aldon Ignitor (pertronix) and Aldon recommended I changed my tach or converted it as link below. It still bounces sometimes? I fitted an RVC. https://www.aldonauto.co.uk/news/tachometer-issues-aldon-ignitor Maybe I can resolve my occasional flick of the needle as well ? R. |
| richard b |
| Richard, I found the same thing with an RVI and Aldon. An RVC fixed it. Maybe you have an intermittent fault, since it works most of the time. |
| anamnesis |
| It’s a Smiths dial RV1, accuspark distributor, inter motor sports coil The dashboard is from a 1970 car |
| R Deans |
| More info---1970 dash but what year is the car and is it neg.or pos. earth- Is the wiring coming from the dist. red/black or black/white Where have you got these two wires hooked up to |
| William Revit |
| 1970, it'll be neg earth. Accuspark recommend converting to an rvc rev counter I believe. I have found several posts from Spridget and Mgb owners, who say the same thing, after fitting an accuspark with an rvi rev counter. But if there is a simple wiring change, then that's great news. What is it? |
| anamnesis |
| Edit: Meant to say, 1970, it'll be neg earth, won't it? 1968 change point? |
| anamnesis |
| depends an. he says 1970 dash-- could be in anything and i'm not about to spit stuff out until i know what the deal is and have ALL the info. |
| William Revit |
| depends an. be patient. he says 1970 dash-- could be in anything and i'm not about to go spitting stuff out until i know what the deal is and have ALL the info. You can't just jump in and say do this or that and end up with smoke when someone is relying on you to give good info. |
| William Revit |
| Anam we have fitted a lot (dozens) of electronic ignition kits to RVI cars and never had a problem with the tacho, so there is no inherent issue with the type either. Kit we use is the magnetronic one, but I suspect it comes down to how it is wired. We always take the module power feed from the fuse box, not the coil as is often the case, so that may or may not be it.
We also have an RVI in the K midget (VVC with wasted spark), joined up on the same principle as original: the power wire to the coil pack does a loop in the tacho en route to the coils. And the RVI is fine with the 123 in my B. Ultimately RVI is counting the frequency at which the primary current is switched off to generate each spark. Provided the ign power wire isn't also feeding other functions the current will drop briefly to zero(ish) each time a spark is required and if that happens it won't matter what is switching it off and on each time. All the tacho sees is the current collapsing and thus counts another half turn of the crank. The principle is the same for RVC, only in that case the tacho is responding to the voltage dropping to zero rather than the current. |
| Paul Walbran |
| EXACTLY-Paul, the unit acts as a condensor accross the coil- red wire from the dist.to the white or green on the fusebox is the normal neg. earth fix--------but i'm not going to suggest doing that if it's pos earth with an accuspark because it can be either a pos earth unit or a neg earth wired up to a pos earth system ,, wired up completely different to each other We still need to know what he's got----------------to reduce the smoke |
| William Revit |
| Yep yep. Depends it seems on the electronic ign' kit used. I tried altering my wiring, various ways with an ignitor; it didn't work. I plugged in an rvc andit did. I haven't tried an accuspark distributor, but the maker says it won't work with an rvi. They don't suggest wiring changes. Why not? Maybe you should tell them. Accuspark say must be neg earth. Since RD, hasn't reported smoke, his must be neg earth; I would guess/assume. Trying to find their advice re rev counters.  |
| anamnesis |
| The RVI type counts current pulses which with conventional points would be the order of 4 amps (12v divided by 3 ohms) which is the switching current. The white ignition wire will loop around something that forms a half loop which feeds the electronics. There are 2 types of RVI, one with external coupling loop and the other with internal coupling loop. This is with my experience from many years ago when I installed an electronic ignition and my car had an RVI with an external coupling loop. I reasoned that when the electronic ignition was fitted the switching current would be greatly reduced. So what I did was increase the number of loops around the coupling loop, after all it is just a loosely coupled transformer. I cut the white wire near the RVI, inserted a length of thinner wire and looped it around the coupling loop about a dozen time before completing the circuit. The RVI then worked properly.
Rob |
| MG Moneypit |
| Ok so it is a negative earth the body is a 1500 and as far as I know the wiring is from a 1970 car but the wiring has been hacked and changed so many times it does not have a ballast thing. The accuspark was wired as per instructions from accuspark The car runs really well only issue is the rev counter |
| R Deans |
| Rob's thinking about the current being reduced with electronic could be a clue. I am guilty here of ASSUMING - something I tick the workshop boys off for doing. In this case I have assumed the coil current remained largely the same on the grounds that a current change of that magnitude is needed for the coil to do its thing. But maybe it doesn't or maybe a different coil is fitted. Reduced current would certainly affect the tacho unless something like Rob's solution is done.
I can feel some exploration with an ammeter coming on. BTW, my statement above about the voltage fluctuations with RVC was back to front... The voltage at coil negative is zero when the points are closed, rising to (nominal) 12 when open. |
| Paul Walbran |
| Yep, that all makes sense. The rev counter sits in series with the primary winding of the coil. Not much current needed to switch the coil on and off by the electronics, but is more when on points. The coil is a voltage transformer, so it still works fine. But the rvi with reduced current isn't stable, because as Rob says, the loop doesn't now carry enough current to induce enough voltage to trigger the tacho reliably. Additional turns does the trick. I hadn't thought of that. Clever and simple. I'm just simple. 🤣🤣
I wonder if the aged components in the rvi also contribute to failure, notably the capacitor in this pic. I replaced one on my Sprite and my Capri, but haven't tried electronic ignition since doing that.  |
| anamnesis |
| An -you're still making a big issue out of nothing, and to be honest confusing a simple issue-- the tacho doesn't need modifying it's a simple wiring fix.
read below and learn R.Deans Now that you've supplied the needed info.--thankyou. If you've wired it as Accuspark have advised, the two wires from the distributor (1 red and 1 black) will have been fitted - red to the pos. terminal on the coil and black to the neg.terminal of the coil With it wired like that the distributor module is getting it's power source from the same wire as the ignition coil and effecting the tacho signal--(A bit like having a shockabsorber in the wiring smoothing out and reducing the signal) The fix is--Take the red dist. wire off the ignition coil and instead of getting the power for the distributor module from there make up an extension for the red wire and connect it to the ign. source at the fuse box instead---I'm not going to say 2nd from the top or second from the bottom as many of these fuseboxes have had their position or wiring altered over the years--so--There is a fuse in the box with white wires on one end of it and green on the other---you can connect to either end of this fuse,it doesn't matter, whichever end has a spare blade to hook to---white is ignition from the switch and green is the same but fused. Cheers willy |
| William Revit |
| Talk about stringing it out to an anti climax Willy. I was expecting "learn" something, that I and others haven't already tried. You could have said all that at the start. There was no reason to wait 🙂.
And with respect, that's a tad condescending Willy. Perhaps you didn't mean it to be so.😁 What you have suggested, I and many others have tried. On mine it didn't work. A quick search across many other classic car marques that employed our type of rvi rev counter, reveals that others have tried the same, and found it didn't work for them either. But it DID work on some, and there has to be a reason for that. That's why I mentioned the capacitor swap, as the rev counters are old, and components degrade. And as they degrade, small changes, like the draw from an additional current drain, connected in parallel, can affect the rev counter. Not a shockabsorber Willy. The correct term is 'shunt'. It's an alternate path for current to flow. Your connection suggestion, is electrically IDENTICAL to the original connection isn't it?. It's the SAME power source. It just draws power(current) from further up the line. Effectively, the same thing can be achieved by replacing the ign' wire from the switch to the pos' on the coil, with a larger csa wire. Larger csa, equals less resistance, which in theory would allow greater curent flow to supply both the rev counter and the additional current demand. And it may well work. And simple if it does. Definitely worth a try. Hope it does. But it doesn't work on most, it didn't work on mine. Rvc's replaced rvi's for a reason I guess. And it's the suggested fix by accuspark and others for this problem. So if the connection change doesn't work, or results in an unstable rev counter needle, it's well worth understanding the whole picture and the remedies, even if some people are confused. 😀. |
| anamnesis |
| "Talk about stringing it out to an anti climax Willy. I was expecting "learn" something, that I and others haven't already tried. You could have said all that at the start. There was no reason to wait 🙂".
There's always a reason to wait until you know all the facts an. What if it had turned out to be a pos. earth car, I'm not here to be judged ,I'm just here to help if i can. "Your connection suggestion, is electrically IDENTICAL to the original connection isn't it?. It's the SAME power source. It just draws power(current) from further up the line. Effectively, the same thing can be achieved by replacing the ign' wire from the switch to the pos' on the coil, with a larger csa wire. Larger csa, equals less resistance, which in theory would allow greater curent flow to supply both the rev counter and the additional current demand. NO- you still don't get it. it's not identical in capitol letters to the original connection---The original ignition wire from the ignition switch goes through the tacho where the tacho senses pulses in the power supply to the ignition coil--and a larger wire as you suggest will do naught " Not a shockabsorber Willy. The correct term is 'shunt'. It's an alternate path for current to flow. " NO I'll stick to 'shockabsorber' to describe what's happening in the circuit with both the coil and ign. unit in the same circuit---using the word shunt is incorrect--a shunt is a low resistance connection between two points Basically- originally The tacho senses the yes/no signal at the ign. coil as the points open/close--It's a current draw measurement similar to an amp meter---it's a wave of yes no yes no yes no of current needed Now--with an electronic module the same signal is still there yes no yes no--BUT--with the dist module power supply connected directly to the ign coil things change--While the coil is in dwell mode(same as points closed) the current draw is the same as a points system but when dwell is over the module opens the circuit and fires the coil--all good so far but the tacho doesn't see as much current drop as the module is still drawing power to operate, and the signal now is a smoother less active wave because there's a more constant current draw--smoothed out like by a shockabsorber in the circuit by the more constant current draw--so now instead of the signal being yes no yes no ,it's now yes little yes yes little yes which the tacho doesn't recognise properly By taking the power wire for the ignition unit off the coil and wiring it direct to the ign.source at the fuse box disconnects the unit from the tacho wiring and returns the tacho signal to normal Also your and Rob's understanding that there is less current draw with an electronic ignition is incorrect, it's the opposite actually--That's the whole reason why electronic ignition converted cars start easier and idle better--The dwell period with a typical electronic setup is usually longer than a points system giving a more powerfull spark and drawing current for a longer period between each firing via the longer dwell period means more primary current draw. no less |
| William Revit |
| Willy I did as you said and it doesn’t work just exactly the same I notice if I rev the engine hard the needle moves but nowhere near accurate |
| R Deans |
| Hmm- don't know then--It's always worked for me. You might have to resort to Rob's idea of extra windings around the tacho's sensing post if that worked for him. sorry it didn't work for you willy |
| William Revit |
| Try contacting Accuspark directly? Anam's earlier post has a phone number for them. |
| Jeremy MkIII |
| Did this tacho work in this car with the points distributor prior to going electronic-- |
| William Revit |
| From the first post Willy.
"I have changed to electronic ignition and now the rev counter is not registering the correct revs -- " I think that means it was working before changing from points. Still not working? Oops. Not a suprise..😉. As I said, a quick search reveals most people have a problem with accuspark and rvi. But I haven't found anyone who did what Rob did. So Rob's idea sounds like the cheapest and easiest solution, as you don't have to shell out for an rvc rev counter, or a conversion circuit board. I might get around to trying mine again by doing the sane thing. Although I have to say, I prefer points. |
| anamnesis |
| "Still not working? Oops. Not a suprise..😉"
Bit sarky an. but then posting a smiley face there is a bit harsh-? --we're here to try and help aren't we. Just for the record, I reckon I've fitted probably 50 or so electronic ignition units/distributors , mostly Powerspark but some Accuspark (including neg earth Accuspark wired into pos earth cars with RVI) and Petronix and never had an issue getting a tacho going- I have an Accuspark in the elan and a powerspark in the MG both with RVI tachs without extra mods and working fine IF RD's tacho has the external sensing block it can be wired up to sense on the neg side of the coil by running 2 extra wires but it shouldn't need that and to be honest I'm over trying to help while you're there taking shots- If I've upset you somehow you best spit out what's troubling you. |
| William Revit |
| No big deal Willy. I suggest you read back over your replies to me.
Anyway, having thunk more in this, I suspect the problem isn't the power connection, but the *triggering* connection, to distributor side of the coil. The electronic ign' almost certainly switches the primary winding of the coil, on and off, with a different shape, and perhaps a 'weaker' strength signal compared to a set of mechanical points opening and closing. An oscilloscope would be useful to look at the current pulses in the rev counter loop, with both the points and the electronic ignition. My 'guess', is that why some rvi work and some don't, is because the rvi's are pretty sensitive to begin with. Then the internals, like capacitors age, and the rev counter becomes even more sensitive to seeing the exact pulse shape and strength it needs. There were 2 generations of rvi. Perhaps the early ones (1st generation), whose components having aged/deteriorated more, don't work as well or at all with electronic ign'. And perhaps the 2nd generation, being younger, and having less aged components, and a slightly different internal circuit, do work more frequently or all the time. I have the early first generation rvi. And changing the power to the coil definitely doesn't make it work. But if the signal to switch on the coil is weaker with electronic then the extra loops Rob put in, would as he said boost that signal and trigger the revcounter. |
| anamnesis |
| Accuspark don't publish the electrical spec of their ei module. And I can't get accuspark technical in the phone to ask them. You have to email them.
But you can get POWERSPARK technical on the phone. Most people say the ei modules are the same as accuspark. So I asked the powerspark tech' if they were familiar with the problem of rvi's. Yep they are. They too suggested try taking the power from another point. But acknowledged it rarely works. I also asked, what is the current draw when the ignition module is powered? The answer is only 20ma. With a 3ohm coil, at charging voltage, coil current is circa 14/3= 4.6 amps. At 12 volts it's 4 amps. It doesn't seem to me that 20ma additional current, on circa 4amps through the ign switch-rev counter-coil wire, will make much difference. But I guess if the rev counter is THAT sensitive already (because it's old), maybe 20ma does make a difference, and on SOME rvi rev counters, it's enough to cause a problem. Hence, removing that additional 20ma from the rev counter, wire does the trick for some, but not most according to various posts across the internet. The trigger wire, normally switches the cb side of the primary coil winding to earth via points. It does the same with an electronic switch. It simply switches the primary side to earth to energise the ht. But the speed of the switching is greater/cleaner with an elecronic switch. That timing issue could be the reason why some rvi work and some don't, because the pulse timing now falls outside what the rvi needs. Again, that has to be affected by the age and condition of the cap's, tranny's and resistors in the rev counter. To me, this again bares out Rob's additional loops solution, because it strengthens the signal to the rev counter. |
| anamnesis |
| The Powerspark chap I spoke to mentioned leading and trailing edge triggering.
---------- And I just found this description of how the rvi rev counter operates. "RVI instruments sensed current pulses in the ignition coil circuit and are designed to work with coil current pulses of 3 – 4 Amps. The primary of the triggering transformer (T) is connected in series with the primary of the engine ignition coil, so that when the contact breaker in the engine distributor closes, the current flowing to feed the ignition coil passes through the primary of the transformer energising the core. When the contact breaker opens to provide a spark to the engine, the flux in the transformer core collapses and appears as a short duration voltage pulse across the secondary of the triggering transformer. Any ignition system that does not permit the tachometer’s triggering transformer to fully energise (saturate), or “de-energise” will adversely affect operation." |
| anamnesis |
| And here's another point.
Simple electronic ign' modules have a fixed dwell. Points based is variable. It may be possible to make adjustemnts to make an rvi work. There may be a small hole in the rear of the rvi case, giving access to the variable resistor in the circuit I posted earlier. R5. Might be worth a try too. --- "RVI tachs expect the ignition coil current pulse to be on and off for prescribed percentages of time. Because electronic ignitions don't turn on and off the same way points do, the RVI tachs are typically unhappy and refuse to work when wired with electronic ignitions. RVC types (because they effectively count voltage pulses) work happily with points and most electronic ignitions." "As for adjustment, there are two main RVI boards, (Gen1 and Gen2) one with one transistor and one with two transistors. They both work by sensing the input pulse via a transformer and using this to fire off a fixed length pulse to the movement. This is known as an astable multivibrator. The movement integrates this output so that the more often the pulses, the further the needle moves.. Both can be calibrated a small amount by adjusting the on board variable resistor. The Variable resistor alters the width of the pulse that the circuit sends to the movement. Some (but not all) tachos have a small hole in the case to adjust the variable resistor. If no hole, you will have to open it up. If the capacitors on board have gone tired, you may run out of adjustment."  |
| anamnesis |
| "Simple electronic ign' modules have a fixed dwell. Points based is variable."
I agree with the 1st part but not that points are variable dwell. You can get them where the dwell can be externally adjusted but are fixed as the percentage, dwell angle, the points are open/closed doesn't change in the systems I'm aware of. I was wondering if the problem might be due to more modern adaptable dwell ignition systems being used resulting in a much shorter charging pulse. IIRC they were originally introduced in the early 1970s and sense the charge state of the coil and adjust on a cycle by cycle basis the point at which power is applied to the coil so resulting in much less coil heating. With a fixed dwell points system the coil has to be selected so it achieves its charge in the short duration the points are closed at max RPM which results at slower RPM in the coil saturating and current flowing which just produces heat but doesn't add any extra energy to the resulting spark. With a adaptable dwell ignition a scope would show a fixed duration coil power pulse regardless of engine speed, a fixed dwell would show a fixed percentage (dwell angle) with varying engine speed. |
| David Billington |
| Yep, I got that a bit reversed David. Oops. Lol.
According to Powerspark,-- "With a points system, setting the points gap sets the dwell to a specific angle – dwell is the period that the points are closed, thus energising the coil. With a fixed dwell angle, the coil is energised for too long at low RPM, and insufficiently at high RPM. Our electronic ignition systems feature automatic intelligent dwell, so that the coil saturdaration is always optimised". Maybe it's that optimisation of dwell, that upsets the rvi operation. Maybe because the components are old, and already working at their limits, and it then doesn't take much to push them over the edge. |
| anamnesis |
| Yes so Powerspark is an adaptive dwell system. I've seen before that there are various single chip solutions available for the detail control with some external components and the power switching device needing to be added. I had wondered if Rob's multi turn sensing wire fix worked by providing a short duration but higher amplitude signal so keeping the RVI circuitry happy |
| David Billington |
| R DEANS. In case you haven't fixed it yet, there's an RVC Smiths rev counter on ebay. 45 quid inc post. At present bidding. 100% feedback seller. From a late mgb https://www.ebay.co.uk/itm/145213392797 And another. 45 inc post or best offer. Good price. Again 100% feedback seller. https://www.ebay.co.uk/itm/145213392797 Actually I rhink the same item, under best offer.  |
| anamnesis |
| David--couple of questions for you if I may-semi thread drift-
#1--Have you had anything to do with the Smiths electronic speedos --I've got a handle on the calibration side, that looks pretty straight forward and behaves on the screen as it should but am having trouble getting the wiring side of it to make sense. Smiths helpline is a dead loss and I seem to get the same bloke every time and it's come to a dead end, Just trying to find someone that knows the unit and what's needed to get a result- #2 With this RVI tacho issue- Because the coil gets saturated for longer at low speed with this electronic ignition unit(longer dwell), maybe the coil having so much dwell isn't getting fully unloaded and not giving the tacho a clean enough yes/no signal because of the residual in the line--Theory, how do you reckon it'd go with the tacho hooked between the coil and the distributor on the neg/switching side of the coil- RVI's can go on either side ,that's not an issue, but I'm thinking there'd be a cleaner on/off signal on that side- It only takes a couple of new wires up to the tacho to make it happen- The other possible problem is the coil itself--A lot of sports coils have very low resistance that might put it outside the spec. the tacho expects to see---or maybe he's been supplied with the wrong coil for the job like a coil for a ballast system. |
| William Revit |
| I tried dis side connection Willy. Didn't work on mine, but def' worth a try as it's so simple. Say he has an inter motor sports coil. So 3 ohm. But I just realised he said he has a 1500 body. Ballast resistor R Deans? |
| anamnesis |
| yeah-- you'd think 3ohm should be ok he said he got rid of the ballast resistor but it could still be a ballast coil (1.5 ohm) if he got sent the wrong one--maybe |
| William Revit |
| I have checked for a ballast resistor before fitting I have been told that the majority of the wiring from a 1970 car I will get a old school auto electrician to look at it as I don’t have a clue and from all the comments it’s not a easy fix for someone with my knowledge |
| R Deans |
| Rather than pay an auto elec', wouldn't it be cheaper to buy the rvc rev' counter on ebay I linked to R Dean? Unless he's a mate and doing you a favour of course. 😁 And unless you are keen to keep an rvi rather than a later one. A 1500 would have had an rvc. |
| anamnesis |
| Willy,
#1 no experience of them, have you a link to the wiring and or calibration details. #2 I don't think you're correct about electronic ignition extending the dwell rather I would expect it to be shorter especially with the adaptive dwell type. I'd put a scope on the circuit and see what it's doing but then I have one. I just watched this video https://www.youtube.com/watch?v=w5auLGtzTqY and at one point, about 7:36 , he put a pocket Velleman scope across the coil to check the waveform and duration, looked an handy piece of kit and about 46 Euros he said. |
| David Billington |
| Brilliant find David. That vid' answers a whole lot of questions. Excellent. How did you find that? |
| anamnesis |
| I used google to search for "calibrating smiths tachometer", it was one of the first of many hits on the subject. Looks like that Velleman scope is no longer available but plenty of similar ones at various price points on ebay. |
| David Billington |
| Yep nice tool. I might indulge myself.
I was going to contact him and ask about rvi and ei. But he skipped to FULL electronic, which replaces the coil, and didn't mention the kind we're having a problem with. But 2 things come to mind. 1. I already thought the state of R5, C2, could be a problem. I've replaced both on a few rvi counters. Here's a C2 I replaced on one of mine. I haven't tried it on electronic ignition since doing this. 2. I now wonder even more if those who have success by just rewiring the power lead, have the 2nd gen rvi, with the single tranny one shot multivibrator. It has a bigger internal transformer and loop. Maybe it's less sensitive to the slightly different signal from the electronic module.  |
| anamnesis |
| Yep, what this comes down to, is with points/condenser, the coil draws more current than it needs to produce the ht to fire the plugs.
As you noted earlier David, it just makes the coil hot, for no gain. But the rvi was designed around this level of current. Whereas, most of the the electronic ignition modules, only pull enough current as is needed to produce the ht, and no more. Not enough current for most rvi's; and that's why they don't work. But Rob found a FREE way round it. There's also an external electronic module you can buy from SPIYDA, that supplies the current signal that rvi's need. It's NOT the rvi to rvc conversion board. You DON'T alter the rvi rev counter. -------------------- "Smiths Tachometer RVI Buffer -542- An external module to convert a voltage signal to current pulses suitable for the RVI tachometers. External Module Converts Voltage to Current to drive RVI tachometers If your classic car has had electronic ignition fitted, the tacho is misbehaving but otherwise is in good condition, or you don't want to upgrade the internals, this will solve the problem. The module is voltage driven and can be driven from electronic ignition or from the coil -ve The module outputs current pulses suitable to drive an RVI tachometer. Note, for long term reliability we recommend a full conversion to voltage triggering for old tachometers as it replaces the old electronics. This kit utilises the original electronics in the current driven tachometers which in most cases will be ( for electronics) quite old having spent most of its life in a hostile environment." https://spiyda.com/smiths-rvi-buffer.html  |
| anamnesis |
| Anam,
"Yep, what this comes down to, is with points/condenser, the coil draws more current than it needs to produce the ht to fire the plugs. " Not the case, the coil peak current is governed by the coil resistance, being an inductor it's a bit more complicated but for the sake of simplicity I'll say resistance, when power is applied across the coil a current will flow but being an inductor it'll fight the change in current so it'll increase gradually until the peak current is reached when the coil is saturated. Whether points or semiconductor switching device the peak current will be the same, ignoring slight voltage drop with the semi conductor. What is different with points is the current continues at the peak current until the points open, that period of peak current from saturation to points open has no benefit and just creates heat in the coil. The adaptive dwell systems just time the application of power to the coil so that it just reaches saturation so the coil isn't powered for longer than required. The energy stored in the coil should be the same. With an adaptive dwell system you can run a lower resistance coil which would likely overheat with a points system. |
| David Billington |
| Willy,
#1 I looked up and read couple of installation notes about those Smiths electronic speedos and would say they leave a lot to be desired. Having the expected input signal level information for the speed pick-up inputs would be useful and example wiring diagrams for 3 wire sensors, 2 wire sensors, and 2 wire sensors requiring an external load would probably clear up the vast majority of problems for users. Document likely written by someone who knows how to wire them rather than for someone that hasn't used one before. |
| David Billington |
| Yep--Thanks
The guy I've seemed to attract tells me a different story every time I ask and seems to be on a mission to sell me a tonewheel and inductive pickup that he just happens to sell as a kit--he keeps talking inductive pickup but it's not. The gearbox/transaxle is a Porsche G50 which has a 6 magnet wheel and a reed switch type sensor 'He' said to put a 50k ohm resistor in the circuit to protect the reed sw. I've had it on pin 5 and pin6 with no luck Asked about needing a power supply on pin 3 and he said NO only with a 3 wire sensor It must need power from somewhere as the reed is just flicking to earth unless the speedo senses earth switching--but it's not I'm going to try and get it going today as it's booked in for rego. tomorrow---will report back willy You might find this interesting as well-- http://www.atomic4.com/dwell.htm |
| William Revit |
| Willy,
What model Smiths electronic speedo is it, I've looked at 3 diagrams and each has differed slightly with differing input options so a data sheet for yours would be useful to look at. A reed switch? that doesn't strike me as a good choice as they have a finite switching life and the sort of thing you'd find on a bicycle speedo but not a car for speed sensing though I have seen them used for water level sensing in header tanks where they'd be more appropriate. Any idea of the type as they can be NC or NO, I can see a current limiting resistor would be needed if connected across the supply but then you'd expect one side of the switch to be connected to the speedo input so it goes high or low as the switch opens or closes, that's where the input requirements for the speedo would be handy. I've designed interface circuits in the past and the interface requirements are known so it's easy, odd that Smiths don't publish the information. The atomic4 information just sounds like the adaptive dwell angle which has been around for decades but maybe with the addition of the ability to rapidly switch the coil current to modulate the effective current without causing the complete magnetic field collapse until the coil firing is required. |
| David Billington |
| David--thanks for your interest.
The reed is naturally open, I agree a reed switch isn't ideal but trying to stick to as many std. parts as possible for ease of sourcing future spares- This is for a GT40 replica Here's what came with the instrument set and apparently that's all they have-- https://www.howardinstruments.com.au/uploads/faq/323/Smiths%20electronic%20speedometer%20post%202004.pdf Have the dash out at the moment to double check connections--back at it tonight--pressure's on for rego tomorrow On rechecking what's there--I mentioned earlier that I'd tried the sensor on pins 5 and 6 but should have said 6 and 7 (sorry if that caused any confusion) Today I borrowed a front hub with an abs sensor and tonewheel on it and hooked it into pin6 and 7 in turn and it operated the speedo on either pin---Happy with that, at least the speedo has life in it Back at it tonight, I'll get back to you later- Cheers willy |
| William Revit |
| " What is different with points is the current continues at the peak current until the points open, that period of peak current from saturation to points open has no benefit and just creates heat in the coil. The adaptive dwell systems just time the application of power to the coil so that it just reaches saturation so the coil isn't powered for longer than required."
I think that's what I meant David. 😉. Because it's that difference it seems to me, that interferes with the triggering of the rvi. |
| anamnesis |
| Willy,
Good you've got some output, I assume you have one terminal of the ABS sensor grounded and the other connected to pin 6 or pin 7 gives you an output. With the reed switch for a guess I'd ground one side and connect the other to pin 6 then add a pull-up resistor from pin 6 to 12V so that should give an input to pin 6 low when the magnet is present and high when not. If the reed sensor is a Porsche part then it should be good, I looked up reed switch durability last night and switching cycles are very high for low current non capacitive/non inductive loads. I came across this for a different Smiths speedo and it has a diagram but quite low resolution but it does seem to have the same sort of connections. https://mossmotors.com/media/instructions/440-600_440-601_440-602_440-603_INST.pdf . GT40 replica, I was involved with one of those in the early 1990s and an enquiry came in from Australia and IIRC the speed bump clearance test was cause for concern as they sit so low as standard. |
| David Billington |
| Back to the RVI.
I found this, which suggests (not directly but by implication) why some rvi's work on EI, and some don't, and explains why Rob's solution might have been needed. "I also come across numeraous tachs with a failed transistor input transistor. The symptom of that failure is that the tach will not work in cars with lower ignition current. One way to diagnose this problem is to temporarily pull the plastic block off tha back of the tach, move the wire a bit and put two or three loops of wire through the metal "U" bracket that goes over the plastic block and put that on the back of the tach. If the added magnetic flux gives you enough signal to trigger the tach, then the input transistor is failing. This could also be an indication of problem with your coil circuit as well, however." https://www.classictiger.com/techtips/motach.html ------------- Or it could be you swapped to electronic ignition, AND or have age related failing components. It's a pity we can't find an ORIGINAL technical paper describing the Smiths RVI rev counters, and their exact signal requirements. |
| anamnesis |
| Willy, Having let your instructions doc sink in a bit more the details for pin 3 and the associated note suggest to me that they have provided internal resistor(s) on the input(s) which are normally unconnected but when pin 3 is connected to 12V or ground act as a pull-up or pull-down resistor. I looked at their product downloads page and all that is there are the same docs scattered about the internet so no help there. |
| David Billington |
| David
We've had a win--in the early hours this morning, there was celebration and a happy owner. Last night I reread the brief instructions and like you decided there needed to be a load on the reed for it to sense anything They (being the instructions and the helpdesk fella)say not to connect to pin 3 but further along in the instructions it says connect 12v for a npn and earth for that same pin for pnp Having the reed switched to earth makes it npn so I put 12v on pin 3---measured the output at pins 6-7 and they both had 5v--got my fused ampmeter and measured to earth and got 60 miliamps so definately a sensing wire only----put it all back together--nothing I swear I had a full head of hair before we started building this car Anyway on reflection, if the sensing wire(using pin 6--tried both 6 and 7 but 6 seemed a safer option) is only drawing 60ma, really it doesn't need the 50K resistor that he said to put in the line to protect the reed--I don't think 60ma is going to do any damage to the reed----reassembled it without the resistor and bingo, we have life---the helpdesk guy, I've decided either doesn't know his product or is there to just sell stuff. Reading the instructions now, it's there but man, it's just so vague and brief--definately needs a better, more detailed explanation for each type of sensor. Anyway, it's going and all back together and goes for rego this morning for the transport guys to pick holes in. With the car itself, yeah it's been a real nightmare getting the compliance stuff done, and ride height was one of those issues-this car has a 50mm dropped floor to make it worse so the ride height at the moment is right up to get the required 100mm underneath, it looks a bit weird as is but as soon as it's through she's coming down---This car belongs to a mate and between it and a V8MGB conversion for him it's gobbled up an unbelievable 18months---It's going to be nice to have it finished with so I can do something on my own car for a change Here's a pic of the thing ready to go an I should have the elan back early next week and will hook it up to the scope and get some pics for you to analyse with the accuspark and with a std distributor, i'm a bit interested myself to see just what difference there is on the coil input circuit between the two being that this is where the RVI takes it's signal from----Is there any particular snapshot you want while i'm at it, i'll be putting the accuspark straight back in and leaving it alone so if there's something in particular you want to see --?  |
| William Revit |
| Willy,
A great result, using the Porsche OE reed?. What is the previous circuit with the 50K resistor, is that in series with the reed switch so ground+reed+50k to pin 6 which likely would just not pass enough current and require a pull-up anyway. 60mA seems higher than necessary but the reed should be OK depending on its specs. I can't think why pin7 would need a pull-up/pull-down as it's for inductive sensors but I can see why pin 6 would as it can take NPN/PNP sensors, although the terminology of a load resistor I find odd as they don't require a load as such rather the sensor when activated pulls the load to ground or supply but doesn't do anything when inactive so needs a pull-up or pull-down resistor to pull the input to the inactive state. NAMUR sensors require a load resistance as they change current in operation and that is sensed as the voltage across the load resistance. Can you measure the resistance between pin 6 and pin 3 as that may be useful for future reference. Sorry a fair bit there but many questions after your answers. Leave the Elan and start a new thread afresh and we can address that as and when. GT40 looks great and they really are low. For the first few months I was working on the project the car was up on trestles so I didn't get a true idea how low they were then one day it went on its wheels and I was like whoo that's low. The local vehicle upholsterer got challenged to make ever lower seats and eventually as per the originals some had blisters in the top of the door to clear head/helmet for taller drivers. I had the pleasure of knowing the late Terry Sanger as he lived locally and who raced GT40s in their day and many other things, the guy had a charmed life and was a wonderful natural public speaker, he'd come along to a talk with carousels of slide photos and just talk about the events around them, the audience was enthralled and he rarely got to the end of the slides before time ran out. For Anam!!! I haven't ignored his valiant attempts to drag the thread back on track. The RVI uses a monostable which made me think 555 timer IC as they can do that and plenty of tachometer circuits using it but basically the same as the Smiths circuit just a bit more sophisticated but the same in principle, that led me to come across the LM2907/LM2917 chips again which are more dedicated to frequency to voltage conversion and some variants have features making them more suited to the vehicle environment so various tacho circuits, this one is dedicated to the Smiths instrument http://www.dummett.net/ime/website/ime/a2z/electronic_tacho_driver_using_LM2917N.html . I've attached the device details/application notes PDF. When searching for that device it came up with the Spyda Smiths tacho conversion board, maybe it uses the 2907/2917. |
| David Billington |
| Hey, I've just worked out the real answer on the RVI! Willy and I have never had an issue despite fitting loads between us. What's the common factor? SOUTHERN HEMISHPHERE 😂😂😂 Electrons probably swirling the other way round to you guys... |
| Paul Walbran |
| Wow, nice gt40. Drool drool. 😪. I can't afford one. Lol.
Yep I've already looked at the spiyda conversions David. And the 555. Spidya gives a lot of explanations in various posts on various forums. But is a bit vague about exactly why rvi's have a problem. Like swapping to a pucker Smiths RVC tacho', it can be done, as can a conversion board, and it doesn't that cost much. But it's so much more interesting to discover exactly why rvi's, or at least so many, won't work with the likes of pertronix/accuspark/powerspark. And the fact that some do. If you know exactly why some don't, surely there is a proper fix for all of them? Is it the age and condition of the capacitors and tranistors and external loop? Do only the 1st generation rvi's with 2 the transistor multivibrator have a problem, and all the 2nd gen rvi's work? There's got to be a logical and discoverable explanation. Not that it really matters, but I love mysteries like this, and then knowing the solution. Willy, love to see your Elan. And yep, if you have the kit, pic's of the waveforms of the pulse shapes and timings/peaks of voltages, of points and condensor vs whatever ignition module you use. There must be something different about either the timing or the amplitude, that falls outside the range of the rvi's that don't work. And even points can vary. You can set them to as little as 10 thou and as much as 18 thou and greater, and the engine will run and the rev counter still work. Further reading from Smiths own service manual, explains -----. "The primary of the triggering transformer (T) is connected in series with the primary of the engine ignition coil, so that when the contact breaker in the engine distributor closes, the current flowing to feed the ignition coil passes through the primary of the transformer energising the core. When the contact breaker opens to provide a spark to the engine, the flux in the transformer core collapses and appears as a short duration voltage pulse across the secondary of the triggering transformer." See enclosed official SMITHS SERVICE INSTRUCTIONS IMPULSETACHOMETER (RVI.1000 SERIES). I'm looking for one for the RVC too. |
| anamnesis |
| And of course pictures of the different pulse shapes/amplitudes/durations, between the electronic modules, when connected at the ignition coil positve, or when connected to a fused live as you've done Willy. |
| anamnesis |
| an.
Sorry I sort of drifted the thread a bit there, but jumped on the chance to chat with David while he was there. I'll get some scope pics next week David, yep, it was ground+reed+50k to pin 6 but now no external 50k resistor-so grnd+reed+ to pin 6 - With the dash back in it's impossible to get at the speedo to measure what's at 3-6, I'd imagine it'd be the same as measuring between ign.power wire going to the dash and the g/box sensor wire---not ideal but without pulling the dash out it's about all I can measure. Paul--You might be onto something there, it's all a bit strange---I reckon R.Deans is sitting back thinking he started a storm---hope he comes back with an answer to how he gets fixed up. The only one I ever had an issue with was on an E Type and powering the module direct fixed that--apart from that never had an issue---weird |
| William Revit |
| I don't mind Willy, not my thread. And anyway, even if it was, I love a bit of drifting too. Lol.
Yep sorry R Dean. But you never know, the drifting might come across a universal rvi fix for yours too. 😁. But I assume if there was a REALLY simple obvious universal fix, Smiths instruments would have made a 3rd generation RVI. But instead they went to RVC, which always works on points or electronic ignition. So thinking on those lines, I telephoned SMITHS, now "Caerbont Automotive Instruments Ltd". After all, they should know right? Nope, the 'technical' bloke I spoke to there, told me to look for a conversion pcb. I asked if they had the old documentation. Nope, the original Smiths binned the whole lot yonks ago. So I asked him about the NEW Rvi tacho's they make and sell. Seems they ain't rvi at all. They are digital, but do have the same dial face with rvi lettering on it for 'originalty' purposes. Ok, so can YOU, tell my why the original Smiths rvi's mostly don't work on electronic ignition modules, and why some do? NOPE. He said I probably know more about those old electronics, than he does. ----Seriously? 🤣🤣. I said thanks, and hung up. Lol If the electrons swirl the other way in the southern hemi' maybe you don't really need to change polarity on pos earth cars Paul. 🤣 |
| anamnesis |
| Reading this with interest on holiday I continue to be amazed by the knowledge on this board
A couple of things going back to Russell’s original post, he mentions that this is a 1500 car but with 1970s fittings and that the ballast isn’t connected - and I’m wondering if there’s a clue in that All of the focus as far as I can see has been on the width of the spike that signals the points have closed/electronics have triggered, no mention of the amplitude, is it possible that an rvi needs a greater amplitude to trigger and that somewhere in the combination of ballast/non ballast, condenser (which if I understand it correctly is a capacitor that reduces the spike from points) there’s now too small a spike , Sorry if this doesn’t help but it doesn’t seem an Avenue that’s been explored |
| timmyk |
| Willy,
If not too much trouble I think your method should measure the p6-p3 resistance, whether the internal circuitry will effect the measurement I don't know. Having it out on the bench would isolate the supply being connected to the speedo also but I understand the access issues. Maybe there will others out there tearing their hair out over Smiths speedos who will come across this thread and find it useful. Anam, I prefer the Smiths circuit diagram in the PDF you uploaded to your normal layout as it's laid out in the more classic 2 transistor monostable layout, see images, so more recognisable for what it is, to me at least. I was thinking about suggesting pulling the transistors and testing them, some cheap DVMs have transistor testers, but other than confirming if out of spec once removed they may as well be changed for new and see if that improves the tacho function. The page you posted a link to indicated the problem getting hold of suitable replacements these days which is why I went looking for the more modern replacement solutions.   |
| David Billington |
| timmyk it was new 1500 shell that was used and all the parts from the earlier car were used on it including wiring |
| R Deans |
| willi I do thin k it has become a bit of a storm and i will get back to you all if i find a solution will be looking at it over the weekend I have a whole list of things that are not right with the car so i will keep you all busy Russell |
| R Deans |
| Russell, it's sometimes stormy, but the sun always shines. 😉. Checkout front wheel bearings. And I can see you already have a toe dipping dangerously close to oil sucking waters in your leaking rear scroll. But that's another storm for another day. 🤣🤣. You could always go back to good old simple reliable points. Many have never left them, and for good and valid reasons. I'm in that camp too. 😁.
But I reckon it's worth trying to get your RVI going by adding an additional loop as Rob did. And you try what Willy said, putting the rev counter between the coil and distributor. Timmyk, yep every aspect of the pulse is being considered. Willy and Paul have the kit to visualise the signal. David I found a few of those transistors too. Not cheap, and not easily replaced by alternates as it may change other characteristics too much. But it has to be remembered that most of these RVI's are working perfectly ok on points. So those that don't work on electronic modules, can't be far off I reckon. |
| anamnesis |
| Seeing if any difference in the pulse might be interesting but as Paul mentioned earlier in the thread and as the Smiths doc description of the operation mentions the tacho monostable is triggered due to the pulse created in the sense transformer by the opening of the points, can't see how that would differ between points and electronic switching but maybe a trace would give a clue. Seeing how the tacho behaves might help so maybe Russell can take a video and post it somewhere. The youtube video I watched the other day, link earlier in thread, was interesting showing the behaviour of a 4 cylinder tacho triggered by an 8 cylinder engine, it rose initially and then faltered as it was triggered too frequently and the monostable didn't run to the end of it time, I wonder if anything like a low ohm coil could cause that sort of behaviour like causing a trigger when the power is switched on to the coil and again when turned off. |
| David Billington |
| David +
I have thought that maybe RD has the wrong ign. coil or a really sporty one with low resistance--it would certainly change what the tacho sees---If it's the 3ohm unit that an. suggested you'd think it'd be ok but we really don't know what's on the car, it may well be an external ballast coil without the ballast- he's only done 30 miles in it so that probably wouldn't have shown up as an ignition issue yet if that were the case-- an. ha ha ha --you rang caerbont , good luck with that, sounds like you got the same fella that I got trying to work out the speedo issue---dead loss and the phone calls from here aren't exactly cheap either, especially when he says hold the line and vanishes for 5 mins to work out what not to say when he comes back. while I'm paying the phone bill to talk. |
| William Revit |
| Thought I would just post this here as well
I see from the two questions I have asked and the answers I am getting that there is a complicated beast and I thought I was going back to simpler times The engine was bought along with the car It has been rebuilt with a fast road cam in it big valve head and who knows what else It has a Weber 45 DCOE carb fitted and does not have the PCV system in place The person I bought it from has no more info on it I have spoken to the original owner who reshelled the car his info was helpful with a lot of things but not the engine which isn’t original As said I have only done 30 miles in the car but so far apart from the rev counter and oil leak the engine is running really well So this weekend I will try a different coil to see if it affects the rev counter Try and unblock the canister on the timing cover And any other sneaky faults that pop up Russell |
| R Deans |
| Hi Willy.
I was actually disappointed that the bloke at caerbont didn't know the answers. 😀. I guess that was pretty naive of me though. These days when you ask to speak to 'technical', you rarely if ever, get to speak to someone who really understands their own product and how it works. Not like years ago, when technical really meant someone with comprehensive knowledge in the subject at hand. Yep, a complete waste of time. Lol. David, --- the pulse that triggers the monostable at the base of TR2, is due to the opening of the points. Yep, that's what the Smiths document describes. "When the contact breaker opens to provide a spark to the engine, the flux in the transformer core collapses and appears as a short duration voltage pulse across the secondary of the triggering transformer. This pulse causes TR2 to conduct--". I've been thinking, again. 😁. I've been concentrating more on the initial current flow that charges the primary in the ignition coil, which also flows in the RVI loop, and induces a voltage at TR2. As you say, if it's the points OPENING, how does an EI module affect that? If the EI module current, is lower than it is with with points as it charges the ignition coil (because they designed the EI module's current flow, to be only as high as it needs to be and no more), then the primary loop windings in the RVI transformer won't build up such a strong magnetic field. And that field, when it collapses, when the EI module switches off the current, won't then induce such a 'high' voltage signal at TR2; and the rev counter won't work. Does that make sense? HOWEVER. I emailed the chap who made the video you linked to. His name is Steve. I asked him for his views. In is his reply to me, he says it's the points CLOSING, and the INITIAL current flow that triggers the monostable. He says, ---------- "It's all about the power that is drawn on the primary winding of the ignition coil. Traditional HT coils draw a certain current when the contact breaker is closed. its that current that loops through RVI pick-up That current induces a pulse in the RPM counter circuitry. As long as that current is sufficient , then the RVI will pick it up. For systems that do not use such set-up or currents that are to low , the RVI will have an issue picking up the pulse. Its hard to state which will and will not , it alll depends on the elements used. Trial and error as they say." ------------- I wish I had all the kit I need to study this. I've definitely got the time. But I'm not rushing out to buy a sig' gen, power supply, and scope. 🤣🤣. |
| anamnesis |
| Anam,
The scope traces seen so far as in the youtube video were voltage traces, what would be better for our purposes I think would be current traces as the magnetic field is proportional to the current and it's the change in magnetic field that's producing the trigger pulse. I don't think the current rises fast enough when the points close but also the magnetic field is changing opposite to what it is when the points open so the generated spike if any would be the opposite polarity and not cause the transistor to conduct AFAIK, makes me wonder what happens when the wire passing through the RVI sense is reversed. I expect Willy will be back on the Elan soon and we'll get more info to think about. |
| David Billington |
| When RVI sensor wire loop direction is reversed is doesn't work. Doesn't kill it, just doesn't work. Rewire correctly and all is well. |
| Paul Walbran |
| Paul, Thanks for confirming what I suspected, not what Russell is seeing apparently as he gets some output but not correct. |
| David Billington |
| Been following this thread with interest sporadically as I'm on holiday driving in the States (Yellowstone right now).
However, I think the RVI switching is quite crude I.e. the induction loop just needs enough of a current change and for enough duration to induce around 0.3v or more in the tach circuit transformer secondary to turn on a geranium transistor. It must surely be the case that electronic ignition doesn't quite produce that required current pulse and the additional turns solution just might solve the problem. This assumes the induction loop is wired the right way round. |
| Bill Bretherton |
| Bill, Blooming heck, auto correct more familiar with flowers than the periodic table elements? |
| David Billington |
| David 😂😂 I'll try again: "Germanium transistor" |
| Bill Bretherton |
| Ha ha. That's ok Bill me old flower. We knew what you meant. Lol. I did wonder why you hadn't been posting. Lucky you in Yellowstone. Is it hot there?
Did I miss something? Was the induction loop reversed somewhere? Anyway, given that Russell's RVI works normally on points (assuming it was accurate), then yep, it surely must come down to insufficient curent in the ignition coil primary, when running on the Accuspark. The question is why. On points, when they close to earth, more current flows in the primary of the ignition coil than is strictly needed to produce sufficient HT voltage. The accuspark still switches the -Ve side of the coil to earth. But I'm wondering if there is a current limiter built in to protect the module. And it can do this because it's design feature is to provide only sufficient current flow, needed to produce a decent HT voltage for the plugs. It wasn't designed specifically to drive an RVI rev counter. And yet, some RVI's do work on electronic ignition. As Willy said, although it didn't work in mine, others, or Russell's, taking the power for the electronic module from BEFORE the loop in the RVI, does make some of them work. Why/How? According to POWERSPARK (same type) the module only draws 20ma. That's 20ma ADDITIONAL, not less current in the loop. Assume a 3 ohm coil, 20ma in 3 ohm is only a .06 volt drop. Are these 'old' RVI rev counters so sensitive now, due to the age and condition of the components, that any small amount of change, stops the monostable in it's tracks, or makes it erratic? If boosting the input pulse with an additional turn in the primary loop in the RVI transformer, adds enough energy to drive TR2 (reading the Smiths official circuit), is that a good idea long term, if TR2 is failing? Or is it a failing C1, the electrolytic on the collector of TR2? -- as it's a known failure point and often replaced. I haven't thunk so much about electronics, and these kinds of 'vibrators', for yonks. 🤣🤣.  |
| anamnesis |
| Well--What an exciting day--not
First off the ignition conversion in the elan isn't an accuspark, it's petronix, but that doesn't really matter I guess, it's the same setup with a ring of magnets driving a hall sensor---same as So- fitted the points dizzy and hooked up to the scope and the scope is now going out for repairs- The voltage selector buttons for the screen seem to be working(lighting up)when i select but the screen doesn't change--It was working perfectly 2 days ago doing a compression test and chasing a vac.leak on a mate's v8 With the thing like it is and the probe on coil+ all I'm getting is a straight line, I can blow that up manually but it goes out of focus a bit-the only real observation at that is that the dwell time is probably getting close to twice as long as it needs be at this speed as we thought might be the case. I've posted the fuzzy pic of that but not much use to us really One interesting feature i did discover though is that the petronix module doesn't switch completely to earth. Fiddling about giving it all a birthday before refitting it i found that the module is in the naturally closed position and when the magnets on the ring pass the sensor this opens the circuit. opposite to how i thought it might be actually. so hooking the module up to a battery with 12v on the pos terminal and earth to the base plate , as the dist rotates the output on the black wire switches 12v 0.3v 12v 0.3v Could it be this non 100% circuit switching that effects the signal the tacho sees----or not--making the current draw signal more of a soggy wave rather than a square wave as you would expect with points closing to earth Anyway when i get the scope fixed we'll progress on, whenever the scope gods allow here's the fluffy blown up pic of the +terminal of the coil at idle (points dist) Really sorry about this, I was quite looking forward to getting into it --- and probably sorrier still when the bill comes for repairs car's going good though willy   |
| William Revit |
| Willy,
The voltage across the coil may look like a square wave but the current doesn't, it takes time for the current to build up as I mentioned earlier, a current trace would be more useful I think than a voltage trace in this case but not as easy to achieve. Some information here showing the characteristics https://www.underhoodservice.com/using-oscilloscope-current-ramp-test-ignition-coils/ . Using a low ohm shunt for measuring the current is usual practice but as the coil is fairly low resistance the shunt would have to be significantly lower resistance to not effect the circuit function. An inductive current transformer won't give an accurate reading as it relies on changing magnetic field and when the coil saturates it doesn't change. It would need a Hall effect current sensor to measure the current based on the magnetic field around the conductor, is that something you have. |
| David Billington |
| Anam I can't see 20mA affecting the tach. It must be that the total current or pulse duration/ shape from electronic ignition is marginal for the RVI so some may work, others don't. Weather in Yellowstone quite variable. Temp varies from 50F to 70F while we've been here, affected by heavy cloud and rain suddenly appearing and then clearing quite quickly. Certainly no heat wave! |
| Bill Bretherton |
| I agree Bill that's what I said before, 20ma is bugger all in 3A, and I can't see how it would cause an upset. But somrthing is. That's a good piece about ignition scope traces.
How about a current probe? Does the scope need a built in function to use one? https://www.tek.com/en/blog/how-can-an-oscilloscope-measure-current That's something I'd read about Willy, that you musn't sit with the ignition on and engine not running, as the ignition module's resting position is closed. So the coil heats up. Hope the scope fix isn't expensive. Pertronix is the same as Aldon, whi h is the same as I had, which didn't work for me Willy. So as yours does work by taking the power from a fused live, a comparison with it using the power from the positive on the coil, might give us an answer of sorts. You know what's just struck me? For ages, everybody it seems, has just accepted that RVI tachometers can be problematic with electronic ignitions. Offered solutions vary from converting the rvi to an rvc with a circuit board, fitting a later rvc tacho', adding another turn in the induction loop, changing the power source, etc. But I can't find ANYBODY who has reported/analysed the exact cause. There are plenty of theories and good guesses. And, conflicting descriptions of how/when exactly, the rvi triggers too; when points open or when points close? So Willy, your mission ---- https://youtu.be/0TiqXFssKMY Consider it a service to mankind. 😁  |
| anamnesis |
| Just have std. workshop fittings David, a pair of inductive high tension pickups for coil/plug leads and all the usual clamp ons and probes--there's an inductive pickup for volts/amps that'll read down to 0.1of an amp but that's on the other part of the machine on the digital readout screen-so probably no real help on this.
Anyway, it's off to the 'lectric doctor in the morning to see if the poor old thing can be bought back to life. I was aware of the current buildup being gradual, that would be reflected in the voltage trace I put up with that downwards voltage curve from the point where the points closed as the coil windings charged then plateauing during saturation ---if I'm reading it right.---I've always been a bit electrically challenged but bluff my way through, I probably should get my brother involved, if he will, he's a fully wired sparky. |
| William Revit |
| Oops, sorry David, it's your link, and you already suggested a current probe. I may be flattened Willy, but side by side comparing points to the Pertronix, may show a difference. |
| anamnesis |
| that was/is the plan an. |
| William Revit |
| Pity you're so far away Willy. What you also need for comparison, is an RVI that still doesn't work if the power source is changed. |
| anamnesis |
| Well -- Having a readup on the petronix and stumbled accross this-
There are three types of Smiths tachs: external current loop, internal current loop, and electronic pulse sensing {late models). Current sensing tachs have a wire loop with two turns that passes the coil current through a pickup at the tach. This loop can be external or internal. Electronic tachs have a trigger wire that connects to the COIL- terminal. MODIFICATION OF SMITHS CURRENT SENSING TACHS Older British vehicles that use a Smith current sensing tachs. The wire from the coil positive terminal to the ignition switch passes through a current pickup at the tach. Installation of an electronic ignition may cause erratic operation of a current sensing tach, due to the higher coil current. Modification of the current pickup to reduce the signal level will usually eliminate the problem. Remove the Smiths tach from the instrument panel. The tach has two threaded studs that are retained from the rear of the panel. Label all wires to avoid errors upon reinstallation. WARNING: Improper connection may damage the tach. Locate the external current pickup on the rear of the tach. Refer to Figure A. If your tach does not resemble this figure but has coil and ignition key wires going to a plug, it may have an internal current pickup. In this case, it will be necessary to disassemble the tach. Once the tach is disassembled, you can use the same general procedure as explained in step 3. Modify the current pickup by removing one loop of wire as shown in Figure B. Note the direction that the wire passes through the pickup. If this direction is reversed, the tach will not function. Recalibrate the tach for best accuracy. Connect a test tach and have a helper rev the engine. Hold the tach in the same position it is mounted (orientation may affect calibration). Adjust the calibration screw on the back of Smiths tach until the reading matches the test tach. 4,000 RPM is a reasonable engine RPM to use for calibration. Please note that older Smiths tachs may vary as much as 500 RPM throughout the RPM range. This variation is not the fault of the ignition system. Reinstall tach in the instrument panel. Check all wire connections. |
| William Revit |
| Willy, Have you got a link to that? |
| David Billington |
| Willy, I expect this is it https://mgaguru.com/mgtech/ignition/pdf/smiths_tachs.pdf . |
| David Billington |
| Interesting Willy. The opposite of what we've been thinking. Russell, are you still reading this? Try what Willy found.  |
| anamnesis |
| I was trying to understand where the 10K resistor was supposed to go if the suggested change didn't work, surely not in the wire going off to the coil as that would prevent the ignition working.
For current sensing there are plenty of PCBs on ebay using the ACS712 hall effect AC/DC current sensing chip which look to be suitable. The chip maker's data sheet says not intended for automotive applications but not why, maybe a suitable transzorb between 12V and ground would help clean up any higher voltage transients on the supply. I'd get one and try it with my scope but I haven't got a spark ignition car running at the moment. |
| David Billington |
| I was on the Petronix site but yes same writeup- I suspect the mention of a resistor in the circuit is for RVC tacho signals as the page initialy refered to both types of tacho---bit confusing that |
| William Revit |
| And thinking more on it drawing more current.
More current flow than with the dead short of closed points? What Powerspark told me about taking 20ma may not be true, or may only be true at 'rest', or may not be the same as fir Pertronix. It could be, that when the module is triggered and takes the coil -Ve to earth, the module itself draws a significant increase in current when turning on what I guess would be a power transistor - hence the heat sink compound used when installing it. That may explain why taking the power soure away from the coil positive, fixes some RVI's. Make sense? |
| anamnesis |
| I don't see why a larger current would affect the tach. A larger induced voltage would still switch the transistor on, say 0.5v instead of 0.3v. Providing the base-emitter current limit is not exceeded that is. |
| Bill Bretherton |
| You beat me to it Bill. I agree. I've been thinking exactly the same thing today, while partially dismantling my Porsche. Ooops. Damned predictive text. 🤣🤣🤣. 😉😉  |
| anamnesis |
| Incoming thread drift-
David Had a measure up of the speedo but impossible to get to pin 3 and it's power supply/wiring comes from a dash relay that i can't access either so unfortunately can only get specs off pin 6 (the sensor pin) Pin6 to earth---- 1.03K ohm Pin6-ign on ----- 0.32 V Pin6 ign on ----- 0.06 amp. to earth Bit of useless info really but until the dash has to come out again that's all i have---- willy On the ongoing tacho subject As i mentioned early on I've never had an issue except for one car that powering the module direct fixed. but reading on an MGB site a fella explained that with the module connected to the coil the tacho senses the coil but also senses the module switching which doubles the signal and puts it outside the window expected by the tach-and by powering the module from elsewhere that is a fix for that-----really weird that some cars are effected by this but some aren't--that's the part of this that i'm struggling to understand and why some still won't work with the wiring reroute donI'll find the MGB thing again and post it here |
| William Revit |
| https://www.mgexp.com/forum/mgb-and-gt-forum.1/tach-issue-with-switch-to-electronic.3477865/ |
| William Revit |
| Hi Willy. One bloke in that thread says, ---
"Your internal loop tacho will be affected, if you connected the power to the electronics to the positive of the coil. Remove this wire and take power from any other white wire. The electronic dizzy messes with the current pulse, that triggers the tacho. Voltage sensing tachos are not affected. " But an increased current, results in an increased induced voltage in the secondary of the RVI triggering transformer, which if anything drives the TR2 transistor on harder. Anothe bloke says the electronic module changes the shape ofthe current pulse, making it less 'sharp'. Pretty much everything we've duscussed, and as you say Willy, still doesn't explain why some work and some don't. I'll wager, if you (can/could) replace all the germanium tranny's and the active components with new like for like, aka an original unaged new RVI tacho, it would work however you connected it. |
| anamnesis |
| Anam and others
As an aside (but may be of relevance), I have an ongoing issue with my RVI on points. I've replaced every component except the transistors. The reading falls as it heats up. I thought it may be the thermistor as I'd replaced it with a suggested value from another forum as the original doesn't seem to be known. So I removed the thermistor but same problem. I have some germanium transistors so I'll try those with and without the thermistor when I have time and the inclination. This thread is of great interest to me - it's good to discuss simpler 60's electronics! |
| Bill Bretherton |
| That's an interesting one Bill.
Off the wall question, -- however unlikely I agree in advance 😉 --- , but have you tried another rvi in it's place? Are you certain it's the rev counter and not something else? As you haven't changed the tranny's, it seems likely(?) it's one of those, probably TR2,as that controls the current flow through the meter. As it heats, it's 'turning off'. But I suopose that argument could be made for TR1 too; esp' as they are both old. As David said earlier, pull them and try a multimeter with a transistor tester. Other than that, how about the movement itself? If you have two tacho's, you could try swapping the guts from one to the other. |
| anamnesis |
| Anam No I haven't tried another RVI but pretty certain it is the tach. Still, it may be worth substituting. My DMM has hfe test so could compare transistors but it may be more about one of them drifting with increased temp. Will have to wait, still away! |
| Bill Bretherton |
| Hi Bill, I have a spare RV1 - the later internal type you can try/have if you want. I think it works OK - its in pretty good condition but I know you can test it out - electronics and me just don't make any sense. Enjoy the holiday ! R. |
| richard b |
| Hi Richard That's very kind of you - I may take you up on your offer. Will be in touch. |
| Bill Bretherton |
This thread was discussed between 27/07/2023 and 10/08/2023
MG Midget and Sprite Technical index
This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.