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 MGB Technical - Intermitent Tachometer

The Tach on my 69 B will at times give a correct reading, then drop to about half of the correct RPMs.
Then, for no apparent reason, pop up to the correct reading after several minutes. For quite a while now I have been able to tap the glass, and find that will often move the needle to the correct reading, or close.

I'm thinking of removing the Tach and trying to replace the diodes. Does anyone have any experience with this problem?

John Layman

My '71 has the same fault. I find if I reach up behind the instrument and push the wires a bit to the left, it is fine for another couple of weeks. There must be a dicey connection somewhere but I haven't got around to removing the tach and investigating.

Mike
Mike Howlett

Almost invariably this is a poor earth problem. I had this happen to mine.

If you can't improve the grounding, just get a short piece of wire put a loop in one end & place under knurled nut at the back of the Tacho where it attaches to body. Find somewhere to fix other end of the wire securely to the bodywork.

Does this answer/correct the problem?
Iwan Jones

AFAIK all gauges should have a ground wire under one of the fixing screws, for the internal lights as much as anything else. If it *does* have a good ground already then it does sound like an internal dry joint. Resoldering the joints would be enough, I have never known a semiconductor diode go intermittent - but there is always a first time.
Paul Hunt

I had the same problem with my 72 - it got worse when I added the pertronix ignition. I don't know for sure that that contributed, as it sat for over a year at my engine re-builder's shop.

I found the article on the internet about SMith's tachometers, but this was for a Triumph of some description and it wasn't the exact same circuit. The article mentioned TWO PNP germainuim transistors, and mine has only one - and I wasn't sure it was the same transistior.

While I was considering replacing this transistor as well as the troublesome capacitor that the article mentioned, I got a good deal on another used tach - but this one was for the later 72 or 73 MGB - the one wire connection to the coil negative terminal instead of the two wire connection in series with the positive coil terminal.

So, I needed to do a little re-wiring.

All it took was simply moving the white wire to the negative terminal of the coil, and running a new ignition feed to the coil - from an underhood source, thankfully.

I did have to make a "double female" bullet adaptor, as the wire from the coil to the dash has the same gender bullet as the tach.

I tried it out and found that the newer style worked better - So I took it apart to clean it just enough to make it presentable.

I found the new style uses an integreated circuit instead of the discrete transistor. It seems to be a more reliable circuit than the older style, but I'll admit that I don't REALLY have enough data.

If I were more of a purist, I'd put the same type back in, but now "it ain't broke!"

I still plan to do more tests about points vs electronic ignition to see if there REALLY is a signifiant enough difference to make trouble, or if my old tach just "gave up."


MJ Jones

I have a '72 where the tach works fine for the first 10-15 minutes of operation but then gets stuck at 3,000/3,500 or 4,000, it will come down if I tap on the glass but will ease back up to 3,000+ when i'm at idle. It generally stays up there until the car is parked and shut off for 1/2 an hour or more. I then find it back at zero. Is this an internal mechanical problem or an electrical problem? Note: I have an ancient electronic ignition from the '70s.
Any ideas? TIA Jay
JT Tracey

Jays sounds more like internal mechanical to me, especially if it stays there when you turn off the ignition.
Paul Hunt

I finally made the measurments I spoke about in my previous message.

Here are the results:

http://www.lorieandmatthew.net/MGBTECH/Tacho.htm


It seems that the only conclusive thing I fould was that the impulse tach (even a weak one) should work better with the Pertronix module than with points.

It remains a mystery whether or not the current transformer tach won't work well with the pertronix, or if it is just old "weak" tachs that have a problem.

So I will leave my impulse tach installed for now.

Sorry if I am beating a dead horse here.

MJ Jones

The reason the Petronix gives a lower current is that it uses a semi-conductor switch which needs a certain amount of forward bias voltage, leaving less voltage for the coil and hence the lower current. I have measured 200v spikes in a convention points ignition system (see http://www.mgb-stuff.org.uk/wn_ignitionframe.htm and click on 'Ignition Teory' and the waveforms graphic), maybe the Petronix spikes and has noise for a longer duration and hence is more visible at a given timebase. I honestly can't see how an impulse tach would be 'better' with the Pertronix than points, it either gives an accurate and steady reading or it doesn't, and if it doesn't with points there is a fault that should be corrected. I can however see that the longer noise pulse of the Pertronix could give problems at the tach. The points system isn't designed to switch the current on and off cleanly, but with the condenser and the coil it forms a tuned circuit that oscillates for a period which causes very rapid and high current reversals through the coil and hence a relatively long duration HT voltage and spark. Without this oscillation the HT voltage pulse is extremely short in duration giving a weak spark. The tach electronics are probably designed to ignore this length of oscillation or 'noise', and increasing its duration may well give a better spark but then start to confuse the tach.
Paul Hunt

Paul,

I found it difficult to summarize, so I may not have been clear. I shouldn't have said the impulse tach would "work better" with the pertonix than points.

I should have said that the pertronix APPEARS to give a greater margin than points, based on my measurments.

I have never seen any problems with an impulse tach on a points system. While I didn't measure as high a voltage spike on the points car I used, that may have been due to a lack of maintence - that car wasn't mine, btw.

The problem I noticed was the Current Transformer tach with the Pertronix system. Would a brand new CT tach have problems? I don't know, as I don't have one.

I theorize that the noise I see on the current trace of the Pertronix is causing the trouble - but without a new CT tach to check I'm still theorizing. I plan to test other CT tachs with my pertronix system as well as replace the components known to go bad in these. (transistor, capacitor, diode.)

I should have surmized that the difference could be a PN junction voltage drop. Does anyone know for sure if they actually use a junction device or a field effect transistor? You'll send me back to my textbooks to be sure, but I think that a FET would not give as high a voltage drop.

Thank you for your help.



MJ Jones

My 1971 B's tach died also, so I replaced the 2 mf capacitor and the transistor with a NTE 126 and it now works properly. Cost of the parts was under $8... The mechanical parts of my meter worked properly... Your meter may be sticking....Try a small drop of very light oil on the needle shaft bearings...
Have a look at http://www.mgcars.org.uk/electrical/body_tach.html
Good luck
Jim

Just thought I would report back on what I have tried based on all of the comments. I went through the ignition system and cleaned up wires and connectors I could easily get to. I also tightened up the ground connection at the back of the tach. This has definitely made a difference. I no longer have dramatic drops on the readout of RPMs. However, I still will sometimes get a reading that is about 500 or so RPM low, and that I can usually bring up to the correct reading by tapping on the glass. This makes me think that the problem is internal, and that I should go ahead and try to replace the capacitor and
transistor and Jim did.
John Layman

This thread was discussed between 12/05/2003 and 28/05/2003

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.