MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

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 - SU Pump

After a recant blog where I extolled the virtues of SU fuel pumps, both electronic and points, my 18 month old points pump packed in. On dismembering, the points were burned, their plastic mounts was very singed and the wires from the coil were devoid of their insulation. I extracted the coil and an area of the masking tape was burned through. There is a small resistor soldered to the copper winding which had separated. The resulting arcing causing the heat. Looks like manufacturing fault to me!
I was also intrigued as to the function of the several 1/4" brass washers loosely arranged around the diaphragm. Are they there to give a louder tick? One might expect them on the electronic pump for that purpose, but the points already "tick"!
Allan Reeling

The experience also reinforced the need for an installed "back-up" pump, which fortunately I had plumbed in 6 months ago. Consequently at the flip of a switch I was up and running again.
Allan Reeling

The burning sounds like the result of the points sticking together, which causes a continuous 7 amps to flow, rather than the normal average over time of 1.5 amps.

The resistor was originally the only form of spark suppression - it helped absorb some of the back emf from the winding as the points opened at the expense of additional current when operated, before the capacitor was added. If that became detached then the points would arc more. I believe without capacitor the pumps were mounted in the engine bay of earlier MGBs and points cleaning/replacement was a routine task. Incidentally Clausager writes that the capacitor was replaced by a diode to prevent the points 'buzzing'. That should read 'burning', either what he was reading was illegible or it's a printer's typo. The diode is far more effective at absorbing the back emf and hence resisting points arcing than the capacitor, but made the pump polarity sensitive. Diodes have been replaced by bi-directional transient voltage suppressors, which are better than capacitors, but not as good as conventional diodes.

The fiddly (to refit) brass washers are the earliest method of spreading the forces on the diaphragm over a wider area instead of being concentrated at one or two points. Subsequently replaced by easier to fit five plastic 'figure of eight' dumbells, then by a one-piece plastic component that basically tied the five dumbells together with struts and is much easier to fit.

paulh4

INTERESTING, AS USUAL PAUL. When you talk about the brass washers being an earlier solution. This pump was bought new no more than two years ago!! New Old Stock?
"bi-directional transient voltage suppressors. Would that be labelled 1.5KE18CA 1322V? That survived, whatever it is!
There was no sign, visible or smell, of the shellac burning off the coil windings.
Ordered another Facet as aback up on the other V8 and will fit a spare, German Blue, pump to replace this SU, which I will now run on the noisy Facet Red top.
Allan Reeling

Paul talked to the 11 brass disks used for diaphragm centering ultimately being replaced by 5 figure devices (progressing first through four plastic spacers, all attached together). Recently (less than 2 years ago if I remember correctly), Burlen Fuel Systems reverted back to the 11 brass disks. Having dealt with all three methods of centering the diaphragm, I have found that the brass disks are the only ones that don't wear out prematurely, or gouge the diaphragm material. My only reservation about the brass disks available today is that they do tend to have rather sharp edges that could start gouging the diaphragm material. The edges of the original disks in the older pumps are slightly relieved - either through wear over the years or (perhaps) having been tumbled prior to being used.

I have to disagree with Paul's statement that the fuel pump was mounted in the engine compartment of the early MGBs. The fuel pump was moved the the rear of the cars very early in the production run of the original MGTFs (early 1954). Cheers - Dave

DW DuBois

Alan, did you buy that pump from Burlen? Silly you should have asked and we would have told you to get the old one properly refreshed. I have never had an SU pump go haywire on me, and that is in over 12 LBC's, 3 wives and 40 years owning LBC's. I have had several of those pumps rebuilt as preventative maintenance. They seem to go at least 100K miles or if not driven a lot, last about 40 years, before you should have them worked on.

The problem you describe sounds a lot like someone added a fuel filter before the pump, and let it get clogged up, causing overheating. SU's don't like pumping against a plug in the intake, but do not suffer from pumping towards a plug. Where is the old pump?
R Owen

The old pump was a real no go, I thoughtII Been on the car since 1973 and standing for the last 20+ years. The new one was fitted on an entirely new system. Definitely no clogged filter, muck in tank etc... The fault was definitely the dry joint i would say. Can't remember where I got the pump and strangely can't find the receipt.
Allan Reeling

"I have to disagree with Paul's statement that the fuel pump was mounted in the engine compartment of the early MGBs."

Typo, I meant to write 'early MGs' but the B snuck in when I wasn't looking.

Interesting they have gone back to the brass discs, it's what they show in their rebuild kits which did rather surprise me. I do seem to recall a chamfer on one edge of original discs.

Capacitors are usually buff-coloured with a wire at each end. Diodes (or more correctly diode/resistor) are usually black cylinders with a red and a black wire at one end. The bidirectional devices can be flat discs with one wire off each side, or black cylinders (much smaller than capacitors) with a wire at each end, as here http://www.mgb-stuff.org.uk/pumpr.htm#quench. That reference number doesn't mean anything to me.

There are already two filters on the inlet side of an SU pump - one is the sock on the pickup in the tank and the other is inside the pump.
paulh4

I recently replaced the diaphragm in my pump. The new one came with the plastic spacer, but I decided to re-use the brass discs.

Dave O'Neill 2

Paul, What I have with that long number is a bidirectional device, as per your site picture.
Regarding the brass washer, here is Burlen's reply to my query;
"The reason we went back to the brass spacers, 11 should be fitted, was that we found that pumps were running cooler with these fitted than the plastic spacers used before".
Allan Reeling

Allan - "Regarding the brass washer, here is Burlen's reply to my query;
"The reason we went back to the brass spacers, 11 should be fitted, was that we found that pumps were running cooler with these fitted than the plastic spacers used before"."

I never thought of that as a reason for abandoning the plastic figure 8 spacers, mainly because when I restore a pump, I run it for 24 hours on my test stand and I never felt that the pump was running hot. That said, the figure 8 spacers were always tight when installed around the armature, which will cause the pump to run a bit hotter than with a looser fitting spacer. my concern is that I have noticed that the figure 8 spacers tend to dig into the diaphragm as they often have sharp edges (the first attached picture shows the sharp edges of a new spacer and a spacer with some elastometer particles on it from having gouged the diaphragm.

The second picture shows a new brass spacer next to an older one. You can see the sharp edges on the disk as compared to an older disk. I have save all of the brass disks and reuse them after having tumbled them to insure that there is no dirt on them and the edges are not sharp. I toss the new ones into the tumbler to smooth the edges.

The string of four lobe spacer that Dave shows in the picture he posted is the ones that are really bad. they are really loose in the space around the diaphragm armature, to the point that after some use of the pump, the inner depression of the armature starts rubbing against the coil core, which causes the pump to become erratic in operation. Those spacers that get tossed into the bin any time run into them. Cheers - Dave D.

DW DuBois

Second picture of the brass disks, showing the sharp outside edge.

DW DuBois

Allan and Dave - is it OK if I use a couple of your images on my web site? Attributed of course.
paulh4

Did you mean me or the other one? ;o)

If me, yes, by all means.

Having said that, I don't see any images from Allan.
Dave O'Neill 2

Paul - Mine are fair game also (there are entirely too many Dave's on the various forums :-) Cheers Dave D.
DW DuBois

Here's a pic of the redundant plastic spacer, if it helps.

Dave O'Neill 2

Paul, I presume you mean pics from another posting. Please feel free. I left the pump coil at Steve Newton's workshop by mistake, will post a pic when I pick it up.
Allan Reeling

"there are entirely too many Dave's on the various forums"

And too many Pauls on some.

Yes I meant the two Daves, thanks both.
paulh4

Here are a couple of shots of the failed SU pump.
On the coil the paper is scorched but not the windings, supporting one of my theories that the resistor (has gone missing) had a dry therefore, heating , joint. The other shows the half of the points and the ore than singed base plate.

Allan Reeling

Second shot of points.

Allan Reeling

"supporting one of my theories that the resistor (has gone missing) had a dry therefore, heating , joint."
While I have had a multitude of burned out resistors (the older pumps have a nichrome wire wrapped around the coil, presenting a nice burn pattern around the paper wrapping around the coil. In the newer pumps the nichrome wire is wrapped around a small rectangular paper containing the scorched mark to one area like the one you show).

Your theory of a dry joint (we have always referred to a bad solder joint as a cold joint on this side of the pond) may well be one cause of that situation, but I have found that most of the pumps that I have restored in that condition has been caused by a pump that has stalled in a current on condition with the power left on as someone tries to find out what the problem.

Once that swamping resistor is burned completely through, the voltage spike becomes high enough to completely destroying the contacts of the points. surprisingly, once the swamping resistor is out of the circuit, the coil will produce a spike of voltage as high as 500 volts (albeit with a very short duration). Like you, I always remove the coil from the housing and if the swamping resistor is compromised I replace it with a 90 ohm 1 watt carbon resistor. Cheers - Dave D.
DW DuBois

Dave, on close examination the "blob" of solder supposedly joining resistor to coil had a very neat, resistor size hole in it!!
Allan Reeling

Given the damage at the points area I also tend towards the stalled pump scenario. Whilst an unquenched coil can produce several hundred volts of back emf when the points open it is as Dave says only very briefly. That will 'burn' the points, but from arcing rather than pure heat. The damage to the pedestal looks more like from heat, and points that have been arcing will develop a spike and pit which can lock them together and cause the solenoid to be continuously energised, and arc-damaged points will exhibit a resistance, which will develop heat when current flows.

SU Burlen say the pump takes 1.5 amps, but that's an average over time when continuously pumping in a test rig. In my tests it actually takes 6 or 7 amps when energised, but should only be energised relatively briefly on a test-rig, and only very occasionally when in use.
paulh4

Allan - "...on close examination the "blob" of solder supposedly joining resistor to coil had a very neat, resistor size hole in it!!"

I doubt that the resistor was a standard carbon resistor with the usual leads at either end. I think that if you remove the tape over the burned area you will get down to a flat card of 'fish paper' with a coil of very fine nichrome wire wrapped around it. In all the years that I have been restoring the pumps, I have never found a standard carbon resistor - it has always been either a strand of nichrome wire wrapped around the coil (with a layer of some type of paper between the coil and the nichrome wire) or the later nichrome wire wrapped around a card of 'fish paper'. If the earlier resistor will leave a spiral burned paper covering. With the later resistor will concentrate the charred paper in one spot, right where the leads are attached (see attached picture).

Regarding the current draw of the pumps, yes the pumps of today are in the 6 - 7 amp level, but that has been fairly recent - I would say from the early to late 90s (approx). Earlier pumps had coil that varied quite a bit, but most being in the range of 4 amps. The very early, low pressure pumps (as used in the T series cars and earlier) only pulled around 2 - 3 amps. Replacement low pressure pumps since 1985 use the same coils as the high pressure pumps, only with a lighter volute spring.
Cheers - Dave

DW DuBois

Dave, It was a standard carbon resistor, wire at each end and that's the only thing missing from the photo on my previous posting, and nothing like your pics. The rest is copper coil wire.
You just have to accept faulty product sometimes. Whatever the cause a pump should last more than 4k!!
Allan Reeling

Allan - "Whatever the cause a pump should last more than 4k!!"
You've got that right. I have, in fact seen some rather strange things in the pumps that I have restored over the years, including some coils that are filled with windings clear out to the the very edge of the end caps - no room for any thing other than the card style resistors (which I added). I have yet to try these coils in any of the pumps that I restore yet.

Was this a pump that you purchased recently for you MG? I am also curious about the pedestal damage that you pictured, particularly what appears to be a rather deep grove where the upper contacts rest. The original pedestals were always a hard (almost a bakelite material that was resistant to heat damage). Recently the pedestals have been made of a different material that is much softer (I can almost cut into it with a knife), rather than the hard material that was previously used. When it first came out, there was a lot of problems of the pedestals actually sagging and the mounting screws sinking into the material. I got several pumps where the pedestals had actually broken into several pieces. Obviously BFS has corrected that situation, as I have not seen pedestals made out of that material anymore. That said, I am still not impressed with the pedestals that have been manufactured. recently.

Are you planning to restore that pump? You can get a new pedestal from BFS, but as far as the coil, you will have to reuse the existing one (BFS doesn't sell the coils) and put a 91 ohm, 1 watt resistor in the circuit. You will have to orient the resistor parallel with the axis of the coil (as in the picture) or else the ends of the resistor will rub against the inside of the coil housing, shorting it out (my main competitor found that out the hard way).

Good luck with the pump and if you need any assistance, feel free to contact me at the e-mail address above. Cheers - Dave

DW DuBois

Dave,
according to Burlen, they seem to think the pump is of 2013 vintage, I bought it 2014 but didn't install immediately.
The pedestal is made of a softish plastic which can be cut quite easily, certainly not the bakelite material I remember.
The windings are unblemished, not even a stain on the shellac and there is about 4 to 5mm clearance all round in the housing.
I will definitely re-build it, it's what floats my boat, as they say, but not entirely sure it can deliver for a 4 litre, mildly tuned V8!
Thanks for the advice and the offer, much appreciated.
Allan
Allan Reeling

Allan - "according to Burlen, they seem to think the pump is of 2013 vintage"
I seem to recall that the really bad pedestals were further back than that, but it could have been one lurking in the back of the pedestal locker. Your description of being able to cut the material quite easily with a knife matches the few pedestals that I have run across. That said, your description of the coil certainly fits with what I have seen in the pumps that are quite new (my problem with trying to date the pumps, even the very recent ones, is that I am dealing with pumps that run from the very old ones from pre war with the brass pump bodied low pressure pumps, through the very latest pumps, so I kind of get lost in the decades.

"but not entirely sure it can deliver for a 4 litre, mildly tuned V8!"
I assume that the pump you have is one of the AZX 1307 type. If so, they will pump 18 GPH (in American gallons). If you need more than that, you will have to get one of the AZX 1400 series, double ended pumps which will supply 36 GPH (at well over twice the price of the AZX 1307 pumps).

If you will e-mail me at the address below my name, I have an idea that I want to discuss with you. Cheers - Dave

DW DuBois

"but not entirely sure it can deliver for a 4 litre, mildly tuned V8!"

At a minimum delivery of one Imperial pint per minute and in practice more than double that i.e. the ability to empty a 10 gallon tank in under an hour you would need a Merlin to defeat it.
paulh4

Paul,
The general opinion about fuel pump delivery seems to be based on max BHP. which is delivered at high revs, near the red line. i.e., under hard acceleration.
20 gals/90 litres per hour is normally accepted as necessary for up to 150 bhp.
200bhp needs 28 gals/127 litres per hour.
Obviously you don't spend much time near the red line, but under fueling at the top end means lean mixture, drop off of power and overheating.
An SU'1307 in tip top condition might deliver 20gals/hour. Mine obviously wasn't tip top, and declining! Our 3.5 V8's and my 4 litre, with huge amounts of torque, running at 1000rpm per 30mph, obviously need to spend little time at the top end, which is probably why I never noticed the pump declining. What I did notice was the acceleration improvement when i switched in the Facet!
Better over supply than under.
Allan Reeling

It's a good point. I've been aware of a torque, power and consumption curve for the 4-cylinder dating back to 1962. The figures for the latter are given in pints per bhp per hour, but converting that to gallons per hour 92 bhp at 6000 rpm needs 8.4 gallons per hour.
paulh4

Interesting Paul! I suppose it could be done with a flow meter on a rolling road, but that might be taking academic interest too far! Easier just to keep to over-supply.
Allan Reeling

This thread was discussed between 14/08/2016 and 23/08/2016

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now