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 - fuel pump blues

My '70 B and I nearly got stranded today due to a fuel pump problem. Beautiful day, car running great,then sudden difficulties. The car ran rough for a minute, then cut out. The fuel pump was hot, and if I tapped on it a little fuel could be seen entering the filter ahead of the carbs. I finally made it home by tapping the pump periodically, then by turning the key off and on when the tapping no longer helped. Turning the key off and on delivered a small stroke of fuel each time, as seen in the filter, then the car would go for a mile or so. By repeating this, I was able to put the car back in the garage (mine).

I just got this car about three months ago, and the PO had to put a new pump on just before I bought it. This makes me suspect other problems, but I don't know what to look for. Any ideas?

Michael
michael

Is this the original SU pump, located inboard of the right rear tire?
Dan Robinson

Michael
My sympathies, I have always had recuring problems with SU fuel pumps, sometimes putting me in dangerous situations, until I replaced it with a Japanese pump.
That much cursed SU fuel pump is the reason why MG's have an undeserved reputation for unreliability. By pass it with a bit of fuel line and put in a Subaru pump (ar any low pressure pump). Check the archives but a '82 Subaru pump, or Honda pump can be installed very easly and quickly up in the engine compartment, out of sight but where you can get at it easily. No more problems ever again with fuel pumps. It even makes the same noises as an SU.
Be sure to use a filter before the pump.
Peter

The pump is an original type pump, located by the right rear wheel. It is not the original, as it was replaced by the PO, so I don't know what brand it is.
michael

Lots of opinions on pumps. Do some archive searching and you will find more than you need, I expect. Have heard a number of times that with SU pumps, the filter goes after the pump to prevent junk from getting to the carbs as the pump itself can pump a few bits, assuming your tank is generally clean. Worth a check to see what condition it's in. You should have a drain plug like my 69 so you can drain tank, pull off the filler tube and take off the access to the sending unit and you can see quite a bit about the inside of the tank condition.


JTB
J.T. Bamford

One problem with the SU is sticking points. It isn't difficult to remove the pump, clean the points and reinstall. Or get new points (inexpensive) and just replace them.
Dan Robinson

You will probably get as many opions about this as there are people who reply on this board. Some like the SU pump because you can work on it and because it is orignal. They have their point of view and I respect it. I feel the same way about a points type distributor.

However, I have gone to an aftermarket fuel pump in my old age. This is simply because I don't like to crawl under the car and fiddle around with the pump and gasoline anymore. I have found a small, low pressure, univeral electric pump to be very reliable. That is what I now use on my fourth MGB. Still, I have no dispute with those who like to keep the SU.

John in Music City
John Lifsey

Hi Michael.

The symptoms you describe certainly suggest sticking points, and are VERY well known to Morris Minor owners.
I understand that the standard MGB pump has an arc suppressor which normally prevents points problems (unlike the standard Morris Minor SU pumps that had no 'proper' suppression).
Genuine SU points are availible in the UK.

HTH.. Don
Don

Check out the ground connection. I'll bet that is the cause of many pump problems.
JLG Galbreath

Hello,

A famous issue about SU fuel pump !!!
Dismantle it out from car. Open the plastic back cover. Check for points condition and verify the diode if there is one inside. verify the proper connections relevant to diode installation ( it must be blocked when poser on and it will consume the extra
switching current at points opening ) If all that is OKAY, check at last for a sticking carb. float chamber needle. ALl this job's a DIY .
There is a link from David DUBOIS available on this BBS where you can find out every informations about SU fuel pumps .
Regards
Renou

Broken Record here......
1983/84 Honda Accord pump. Same noises, similar look, same pressure, easy to mount, cheap, reliable.
Cheers,
David
David

Michael,
I'll bet this is what you are going to do, like all of us did

1-attempt to repair SU, get covered in petrol and crap crawling around under the car. You will get the pump going and think you've solved the problem. You will spend most of saturday doing this. And almost certainly badly gash your finger with the screwdriver trying to get the mutiple of slot head diaphram screws undone.

2-pump will stop again (usually at night, in the rain and probably in a 200mph, don't stop or you'll die, speed freak zone) You may or maynot get the thing going by hitting it (hard, repeatedly and almost certainly enthusiasticaly)with a hammer

3-you will conclued that the points need replacing and be mildly shocked to find that they cost about $60.00. Once again saturday will be spent in a haze of petrol and grease. You will learn about importance of winding diaphrams in and out by the right (initially wrong) number of turns to get the "kickover" right. You will think that you've mastered the fel beast.

4-Not long later pump will give a repeat performace as per no. 2 above.
5-It turns out that the coil has become unreliable, it wasn't the points after all. Regrettably replacing the coil involves a two ton press and you can either take the Monster to a professional for a rebuild, or buy a replacement, or leave it as a decoration and use a Subaru (or Honda) pump.

Have fun now.
Peter

I had a similar problem with my 70B recently and it turned out to be the pipe that connects the two carburettors and runs to the canister. The brass pipe on the front carburettor was clogged. Once I cleaned this out my car started right up and ran fine.
Kevin

Peter - I have heard about coils going bad previously, also someone from OZ (possibly from you). I no doubt that you have seen this problem, but in 25 years of working on SU fuel pumps, I have never seen a bad coil in any pump that I have worked on (now that I've said that, i'm sure that I will see severas in the next few months). What I have seen quite regularly (particularly in the AUF 300 pumps used in the MGBs) is that the internal swamping resistor is burned out. This is the result of clogging of the fuel line to the inlet of the pump, causing it to stall in a current on condition. If the power is left on for any length of time when the condition exists, the swamping resistor (very thin strand of nichrome resistance wire inside the coil housing) overheats and burns out. This is also why I don not advise putting a filter on the input side of the pump as they are more apt to clog than the internal filter in the pump. Once the swamping resistor is burned out, the effectiveness of the arch suppression circuit (of which the swamping resistor is part) is greatly degraded and new points wind up burning very quickly.

In Michael's case, since he is able to get some pumping action by turning the ignition on and off, his pump has failed in the above described condition and if one was to remove the coil from the housing (as I do when restoring the pumps), they will find that the swamping resistor is toast. It will do no good to replace the points in this pump as they will burn in a very short period of time and he too will be cursing SU fuel pumps.

On the subject of the Japanese clones of the SU pumps (or even the German or New Zealand clones), all of which are really good pumps, they last about as long as a new SU pump (also a really good pump). I had a Mazda truck with the Japanese clone in it and the pump in gave out in about the same time as SU pumps give out. Even the little square, run all the time, make a lot of noise Fawcet pumps die in about the same amount of time.

If anybody is interested in my SU fuel pump services, a write up of what I do to restore the pumps can be found linked at the end of my article on SU fuel pumps that can be found on Les Bengston's web site at:
http://www.custompistols.com/cars/articles/dd_su_fuel_pumps_101.htm

Peter - Do you suppose the coils going bad has something to do with living in the southern hemisphere and the magnetic field coming from the other direction?
Naw, I didn't really think so either ; )
Cheers - Dave
David DuBois

Dave,
It 's all right , your SU fuel pump experience is obvious. The really annoying point comes from the necessity to crawl under the car to withdraw it for service. May it be replaced either behind the dash or in the front boot successfully ?-regards-
Renou

"The really annoying point "

Ho ho.

The SU for the MGB is designed as a pusher. Pumps on earlier square riggers were pullers, were in the engine compartment for easy access, which was just as well as the unquenched points had to be refaced and replaced on a routine basis. On rubber bumper MGBs the points end of the pump sticks into the boot making it (relatively) easier to get at the points. Being above the tank it is also possible to change these without getting any siphoning. Since access to these pumps is usually only a once in ten years (or much more for low mileage cars) it may not be worth the fiddling about to move a chrome bumper pump into the same position.
Paul Hunt

According to the spec sheets in the Burlen Fuel catalog, the AUF 300/AZX 1300 series pumps can be mounted up to 18" above the bottom of the fuel tank and anywhere from 6" above to 48" below the level of the carburetors. Based on that information, I would have to conclude that they could be mounted in the engine compartment as the LP series pumps were in the T series and earlier cars (behind the dash might be a bit noisy in the passemger compartment, not to mention the gas lines being routed into the passenger compartment). That said, I have to agree with Paul, that is a lot of fussing around for something that occures once every 5 to 10 years (of course, it is always on a cold rainy night in the middle of heavy traffic).

I have dont two things on our MGB to make dealing with the fuel pump less odious. Firts, I installed a shut off valve between the tank and the pump so I don't have gas dripping off of my elbow when I have to remove the pump. Second, I permenantly installed a back up pump, plumbing it in series with the main pump and putting a single pole, double pole, center off switch to switch between the two (not real good advertizement for the guy who restores pumps is it?). I can switch while on the road and the switch doubles as an anti-theft device of sorts. This way, if the pump does fail, I don't have to mess with it until the car is in our nice dry, warm garage.
Cheers - Dave
David DuBois

Dave,
What type of backup pump did you install? I just so happen to have a square, solid state pump that I purchased from Moss about 5 years ago for my '74 B (which I've since sold) & never had the chance to install.

Whether or not I get the SU rebuilt or replace it with a new one, I want to have the backup in place. Where to mount it & how to do the wiring would be the next questions that I would have. I'd like to have the toggle switch in the cockpit, as well.

Thanks,
Gary
Gary NJ

Gary - See your e-mail. That little square, run all the time, make a lot of noise Fawcet pump will work fine as a back up pump. Among other things, once you have to use the back up, that little pump will make enough noise to spur you on to getting the primary pump taken care of so you don't have to listen to the back up any more ; ).
Cheers - Dave
David DuBois

I can't help noticing that a warey distrust of the Dread SU is the ongoing theme ;)

I'm turning Japanese, I'm turning Jananese..I really think so..
Peter

Just to second JLG's comment, do check the earth. I got stranded in exactly the circumstances decribed of fast road with soft verges, at dusk, in rush hour, raining, on a bend. Fuelled with more than a little adrenalin, I bludgeoned the SU with the largest spanner to hand until my arm ached and my knuckles bled. Seeing an uncompromising looking truck fast approching I dropped the spanner and headed for safety. As I did so I heard the pump chatter into life. The spanner wegded between the pump and bodywork and formed an good earth.

Come to think, on the SU debate, at least you can work on the pump on the kerb side of the car in the States. Sounds like luxury. ;)
Steve Postins

Pete said "I'm turning Japanese, I'm turning Jananese..I really think so.."

It's slang for going gay :)

BTW, everyone complains about the noise of a Facet pump but w/the engine running you hear absolutely nothin'. I've used one in various cars and other than waiting for the bowls to fill with fuel prior to starting I never hear it.
Mike MaGee

If you don't hear a Facet type pump on starting be prepared to walk home after the SU's are empty. You can go .8 miles if you only make two stops. The problem was in the bullet connecter ground circut in the boot so the type pump didn't matter. With pump grounded to battery tray since walking home. Clifton
Clifton Gordon

"Fawcet pump ..."

Is that the one that doesn't need a 'tap' ...
Paul Hunt

Excuse me for jumping in, but is there anyway to verify that the fuel pump is starting to go, other than waiting for it to die completely? I'm in the process of running in my rebuilt motor and it's died on me a number of times. It's definately fuel starvation, not electric. At first I thought it was maybe sediment in the tank getting passed through, as it has been sitting for 6 months, but I've noticed that when it slowly dies, if I switch off the ignition and turn it bsck on after about 10 secs she fires right up as if nothing had happened. So reading all the info above, I'm wondering if the pump is starting to go, but I don't want to replace ie. spend money, if not neccessary.
Ken
K.G. Martin

Great idea about the shutoff valve. I received Dave's pictures & was wondering how to get the steel fuel lines & the valve connected?
Gary NJ

Gary - I marked the line going fromt he tank to the pump, then removed it and cut a section out of it, soldering each end into a male pipe fitting that would fit in the threads of the valve (they are going to be either 1/8" NP or 1/4" NP and may be straight or tapered depending on what valve you get). I then screwed the whold mess together and reinstalled the pipe. You can get the valve (1/4 turn ball valve from any industrial supply house such as Grangers or a good hardware store that has a good selection of plumbing items (I am thinking of something other than Lowes or Agent Orange, although they may have something). The great thing about the valve is no more gas dripping from the elbow when changing a pump and it can serve as another anti-theft device. Cheers - Dave
David DuBois

K.G. Disconnect a hose from one of the carbs (there should be a spurt if the ignition has been on recently) and direct it into a container. Turn on the ignition and it should pump at least 1 Imperial pint per minute.
Paul Hunt

Mike
Urrck, ugg ummm, not around here it doesn't.
The song, the song!!
Not that there's anything wrong with that but..no
Peter

Ok, I removed and disassembled the offending pump. It has some type of solid state drive rather than the points shown in the manual. Also. there was no inlet filter, the fuel flow is unobstructed, and and everything looks clean. Apparently there is some flaw in the pump so it only strokes once when the key is turned on and off, but it was not a clogged inlet. Again, this pump is only 4 or 5 months old and the car has not been used much.
Any recommendation from here?

And no, i'm not turning Japanese (at least not yet). By the way, conversion look like a pain, with piping and wiring changes...

Michael
michael

This might be a problem with the kick-over. However I'm not familiar with the pointless ignition set up you've got. People buy this so that they don't have to keep replacing the points, lay full length in the mud, bash SU pumps etc. It's supposed to be better. This may or may not work with you however with the POINTs set up this is what you do. If I get this wrong someone correct me, maybe David, he appears to have the necessary supernatural powers to get these things working reliably long term.
The diaphragm needs to be turned in or out the right amount to properly tension the points/set the stroke. The diaphragm is attached to the spindle which runs up through the body of the pump and a screw thread sticks out dead center at the back of the pump and is screwed into the mechanism of the points. You rotate the diaphragm and the screw puts more or less tension on the points set up. Screw in and the pump becomes less inclined to kick over (with a shorter stroke), screw out and it becomes more able to fire up (longer stroke) , to the extent that it may continue running pointlessly (ho ho) even though the carbies are full. Try winding it out a couple of screw holes worth (120 degrees? for starters.
You do this by unscrewing the many slot headed screws that hold the two halves of the pump together , you will need a largish screw driver (band-aide on standby). Thoroughly clean up the pump around this area before you start, you don't want any dirt getting in.
Gently loosen the diaphragm from the edges, it tends to stick to one half or the other so a delicacy of touch is required, or you'll be buying a new one.
Essentially it's trial and error from here on. If that doesn't work, screw out some more, then some more and it THAT doesn't work, then it's likely your pointless points are ..well err...

FYI
Above the passenger foot well, inside the guard and accessible from rear of the the engine compartment, is a small "dead" space. If you were to wrap a piece of foam rubber around a, for example, '80's Subaru pump, it would just fit in there, accessible and out of sight. The tight location and the fuel lines will hold it in place. The Japanese pumps are pullers as well as pushers because in the subaru they occupie the top rear port side of the engine compartment, which is where you'd look if you were at a wrecker for example.

Best of luck.
Peter

Michael - I suspect that you don't have a SU fuel pump. Is the body plastic and are there two spade lugs on the end cap for attaching power and ground? If so, you have one of the Auto Pulse pumps (I believe that is the name of the pump) made in New Zealand (no comments Peter). These are good pumps, but they are a throw away due to the way they are made. Also the plastic tends to develope cracks around the banjo fittings over time. The other posibility is a German pump with the brand name Hugo, which is another throw away as they really can't be fixed. If you have a digital camera, take a picture of the pump, convert it to JPG format (to get the size of the file down since I am on a dial up) and send it to me. I may be able to get something going for you.

Peter is correct on the setting of the points in the SU pumps. He is also correct about it sometimes being a hit and miss procedure trying to get the points adjusted correctly and the diaphram set properly. It is particularly difficult when one has to continually install and uninstall the pump to test and readjust and just getting it to run with no load on it, doesn't guarentee that it will keep running after installation.
Good luck - Dave
David DuBois

Copy of something I posted elsewhere-
I have been using Walbro pumps lately though, these are the round ones that you can get from JCWhitney among other places. I think the square Facett (Fawcett?) ones are the same. These have a hollow piston in the line, with the check valves in the piston, driven by an external concentric solenoid coil. I believe the piston is pulled back by the solenoid and pushed out by a spring, just like the SU diaphragm. However, I have never taken one apart, having no failures, and I can't find the Walbro literature. It is possible that there is no spring and the piston is driven solely by the solenoid in both directions.
I find the Walbro design very elegant and troublefree. It is solid state, totally sealed, has two wires so is good for pos or neg earth, small (1/2 or less of the SU), cheap($35) and reliable. If mounted properly with outlet upwards, it is totally immune to vapor bubble trouble. It is much quieter than the Facett and is easily mounted in rubber if desired. Its a great backup even if you want to keep your SU or whatever; if it is left off, the other pump will just pump through it. Walbro is a major OEM supplier for automotive pumps.

FRM
http://www.usachoice.net/gofanu
FR Millmore

Hey Pete, I remember the song very well :) At least in NYC that was the meaning of the term.
Mike MaGee

FR,
It's "Facet":
http://www.gowerlee.dircon.co.uk/Facetpump.html
;-)

Michael,
Is this your pump?
http://www.mossmotors.com/Shop/ViewProducts.aspx?PlateIndexID=29521
Rob Edwards

In our country we refer to becoming gay as "turning Brooklynese".
Minoru
Minoru Genda

That is my pump. I will be replacing with a rebuilt SU, but I still don't know why it failed in only a few months. It is pointless, the inlet and outlet screens were completely clean, and the lines seem to be clear. Oh well...
michael

Hi all.

Some general observations about switching inductive loads..
(I have no direct experience of 'pointless' pumps, so observations about these are strictly general).

One problem is that the points (or the transistor in the case of 'pointless' types) have to cope with a very high voltage due to the 'back emf *' that occurs when the coil current is broken.
This back emf tends to cause arcing (in the case of points) which can cause the points to weld together, or go out of adjustment due to erosion or go high resistance due to burning.
In the case of switching transistors they often go collector / emitter short circuit, which means that the pump will click just once when the power is applied.
An 'exotic' contact material (eg platinum / rhodium / silver / gold) can help with points, while high spec transistors can help with 'pointless' types.
Spraying the points with a little contact lubricant helps, too.. I know that it can attract dirt, but that need not be a problem if a little care is taken.

A semiconductor diode across the coil does a really good job of 'killing' the back emf, but unfortunately slows down the operation of the pump to an unacceptable degree.

A capacitor (modern name for a condenser) across the points does help, and was fitted to some SU pumps.
(they are small silver cylinders in the ones I have seen).

I have successfully used power mosfets (a type of transistor) to switch inductive loads in the past, it is possible that the ideal solution is to use conventional points to switch such a power mosfet.. the points would not be exposed to any significant current or back emf and a switch could be provided to enable the pump to be run directly from the points in the unlikely event of mosfet failure.

Another option is to use modern suppression devices, such as zener diodes or MOV's (Metal Oxide Varistors) which 'break down' to dissipate the excess energy in the back emf.

*(It is 'back emf' from the coil that provides the spark at the spark plug.)

Is anyone still awake out there.... ?

HTH... Don

Don

My two cents worth,
There is an aftermarket pump actually I got it on e-bay from a company in England. It is an exact replacement for the SU but the points are gone and it's replaced with electronics I am not sure if it's a SU.

cec
Cec

Cec - Burlen Fuel now put out a all electronic SU fuel pump. If the pump has SU on the alloy casting and on the end cap then it is a SU fuel pump.

Don - You are correct about using a diode to quench the back EMF on an inductive load, Burlen Fuel has been using them for a number of years on the SU fuel pumps and as I recall, they have a diode across the coil with the all electronic circuit. The SU pumps have always had a swamping resistor built in to the coil that aids in arc suppression and if that resistor burn out (as from a pump that is stalled in a current on condition), even the diode will not suffice alone for effective arc suppression. I firmly believe that is one of the reasons that people often find that replacement points burn up in a very short period of time. Switching a transistor on and off with the conventional points has it own set of problems. The tungsten points build up a film when they are idle or if there is insufficient current density through them to keep the film burned off. I found when I was trying to drive a transistor with conventional points, that if the points were not adjusted exactly right to provide a wiping action as the opened and closed, the film would build up and cause the pump to fail. I finally replaced the contacts on the points with a magnetic reed switch and a magnet and have not had any problem with the pumps since. As to a diode decreasing the pumping speed of the pumps, I have experienced that in some of the older pumps, but the later pumps don't seem to be affected by a diode (even with the transistor, I use a diode across the coil to quench the back EMF).
Cheers - Dave
David DuBois

Hi Dave.

A while ago I got involved in a problem involving very lightly loaded REED switch contacts going high resistance !....

Did you retain the spring toggle arrangement with your reed conversion ?.

I have some hall effect devices that I might try when time permits, they seem to do the job on modern engine management systems... maybe a crankshshaft position sensor could be modified to provide detection ?.

Don

Don

Don - So far I have not had a problem with high resistance in the reed switches (other than when a base to collector short has developed in a couple of the transistors I use, but then that caused the reed switch to go into the incandesant mode).

I use the original spring toggle arrangemend as the carrier for the magnet. This seems to work quite well once I got an adhesive that stands up to the constant shock of the toggle at the end of it's travel. I had a problem with epoxy, such as JB Weld failing and the magnet coming loose.

I have not tried to develop anything with a Hall effect device since Burlen Fuel beat me to that method and would probably take a dim view if I infrenged on their development. By the way, I have used their Hall effect circuit in a fuel pump that I presently have in our MGB to see how well it works. So far so good. I do find that the Hall effect pump runs a bit quieter than the original pumps with the spring toggle. Someone is sending me a Hall effect pump that quit on them, so it will be interesting to see if the circuit failed or was not adjusted quite right originally.

We are going to be over in England in lat August through the first two weeks of September. If time permits, perhaps we can get together and compare notes.
Cheers - Dave
David DuBois

This thread was discussed between 10/05/2004 and 19/05/2004

MG MGB Technical index

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