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 - SU fuel pump fails when cold?

I'm wondering if anyone has ever had a problem with their fuel pump during cold weather. I usually drive my 67 every day. About a week ago after not driving her for a couple days the fuel pump failed to pump after about 30 seconds when I was waiting to back out of my driveway. A brief rapping on the fuel pump body got me on my way.

I replaced the filters(one between the tank and pump and one right before the carbs.) as it had been a while since they were replaced and were looking a bit dirty. I hoped a dirty filter might have contributed to the problem as when I first bought the car a few years ago I had the car die when the fuel filter clogged up. The PO had installed the clear plastic filter backwards so the dirt was all collected on the inside of the paper and thus didn't look all that dirty until you realized it was on the wrong way.

Anyway, a few days later I went snowboarding and the car ran fine up there and back(85 miles each way.) However after sitting in the sub-freezing parking lot all day the engine died of fuel starvation once again when I was warming it up before the drive back home. Again, a few raps with a hammer got the pump ticking again with no additional problems encountered since. This pump on the car is 2-3 years old and has points.

Now on the 70 I had previously I had a similar problem. The car was fitted with a fairly new(still had points) midget style SU pump and it too failed durring an outing in cold weather. It actually failed several times enroute when I was above the snowlevel. I was so close to the mountain that I kept driving there anyway, stopping every several hundred yards or so to tap on the pump to get it going again. Coming home that time I had to stop to hit the pump a few times as well but it wasn't so bad as I was coasting downhill most times and it would often resume pumping before I came to a complete stop. Once below the snowline I had no problems driving the remaining 60-70 miles home. I had no prior or additional problems with that pump.

Surely the current pump has exibited a certain amount of unreliability by failing once before in my driveway. However It seems that between these two occurences on different cars and different style pumps that maybe there is a problem with SU fuel pumps and cold weather? I have never heard of such a problem, I have heard of fuel pumps oveheating, but most electrical equipment doesn't mind running in a slightly cooler environment so this puzzles me.

Has anybody else had any similar problems or do I just have poor luck with these pumps? I'm thinking of heading out of town this week but will probably invest in an universal pump to carry as a backup before I leave town.

-Jared
Jared Snider

Jared. I used an MGA as a daily driver in Cheyenne winters with a Healey 100 as back up (the Corvette could never be considered a reliable back up to either) with no fuel pump problems. I used an MGB in German winters and some time in Denver during the winter without problems. All of these cars were used to drive during snowy weather, including trips to Wyoming and Colorado ski areas. Hence, my several years of experience indicate the SU pump is reliable in cold weather. You might contact David DuBois about your problem and see what he as to say. Les
Les Bengtson

Just clean/replace the pump points. It is quite possible that in cold weather different components contract at different rates changing the physical alignment of the points such that they fail. It isn't the cold that is the problem, but marginal points. There is no generic problem with SU pumps in cold weather.
Paul Hunt

Cold weather can cause condensation, especially in high humidity areas like the Pacific northwest or the gulf coast, (where I live). Lots of dampness and fluctuating temps could be causing some moisture condensation inside your points cover. Clean them well and make sure the inside of the cover is dry. Should make a difference.
R. L Carleen

Jared - I'm with Paul, new points are in order. I would also look at the existing points for signs of excessive arcing. When you say that the filter on the inlet side of the pump clogged, the pump may have overheated enough to burn out the swamping resistor inside the coil housing, thus making the arc suppression circuit much less effective. The burned out swamping resistor is a rather common situation on the pumps that I have seen and one of the reasons I don't advocate using a filter on the inlet side of the pump, they just serve to get clogged and stop the pumps in a current on condition, causing the resistor to burn out. Good luck - Dave

PS. See my article on SU fuel pumps at: http://www.custompistols.com/cars/articles/dd_su_fuel_pumps_101.htm
David DuBois

Thank you, everybody for such quick responses. It seems that a SU pumps in good condition should do just fine in cold weather and that is not out of line from anything I have been led to believe before. It seems that the marginal condition of my individual pump is the cause of the problem. I will indeed have to investigate the points soon. I don't think the fuel filter really was clogged but I wouln't be surprised if the resistor has been burned out. I would have to take responsibility for this happening.

Since the MkI cars do not have an accessory on position incorperated into the ignition switch I devised a way around this. In order to be able to have the radio work without running the car I installed a switch to cut power to the coil so I would not be in danger of burning out the electronic ignition module. However it it seems that I took no consideration for the fuel pump and as such it has been subjected to a current on situation many times over the last couple years. So it seems that I probably have been the cause of the problem. I will have to remedy this. I hate being the DCO, I'd rather blame the DPO for shody electrical work.

-Jared
Jared Snider

Hi Jared.

This is a bit of a long shot, but just maybe...

The contacts do need to be correctly adjusted to work properly. It is just possible that your points may be on the edge of adjustment, and thermal contraction on a cold day is enough to make them unreliable.

The fact that a mechanical shock gets things moving again certainly suggests a problem with the points.. Morris Minors were supplied with a starting handle, which was usually used for hitting the fuel pump more than for turning the engine... (the SU pumps on moggies didn't have a spark quench).

Don
Don

Don - The moggies fuel pumps had the first attempt at arc suppression, that is they have the swamping resistor. On the early L Type pumps that drew very little current, this was sufficient to keep the arcing under control. As the pumps got more sophicated and drew more and more current, it was necessary to add things like capacitors and finally a diode to hold the arcing down to tolerable levels.

Jared - Take hart, it was not your jury rigged accessory power that would have done in the swamping resistor. Even when power is supplied to the pump constantly, there is no current being conducted through them until the points are closed. In normal situations, where the pump clicks once and then waits for the pressure in the float bowl to decrease sufficiently for the diaphragm to relax and close the points again, there is no current being drawn and therefor, no heat being generated in the swamping resistor. It is only when the pump stops due to a vacuum being drawn on the input side (as when there is a clog in the line between the tank and the pump) that the pump will remain in a condition where the points are closed contenuously and the coil and swamping resistor drawing current contenuously. You can sit and enjoy the radio all day long and not harm the fuel pump as long as there is no clog on the input side.
Cheers - Dave
David DuBois

Hi Folks:

I have resolve the SU fuel pump point burn-out problem by using a simple TO-220 case transistor (TIP -41) for positive ground pumps, a resistor (25 ohms) and diode (1N4003), and a solder lug. Essentially, the points now switch the transistor on and off. Consequently, the points are now switching only a few milli-amps instead of amps. No more buring of the contacts. For negative ground cars use TIP-42. All these parts can be found at Radio Shack for few bucks. The circuit schematic and how to do the pump modification, is delineated on the Jag web page. I only have copies of the design. Note: It works great!!!!!

Rich Boris 67 B roadster, with positive ground.
Rich Boris

Rich - I wrote the article that you see on the Jag web site and have been trying to get them to take it off of their site due to errors in it. First, remove the 25 ohm resistor and replae it with a straight piece of wire. You will find that the pump runs better without the resistor. Next, you may find that the pump will quit pumping after a period of time (this doesn't always happen, but it happens enough that I went to a different method entirely to drive the transistor), acting just like the points are burned again. In fact, cleaning the points will restore the pump to proper operation for a short period of time. What is happening is that the points develope a film over time that insulates the contacts because there is insufficient current density to keep the film burned off. This is particularly true if the points have been replaced with a set from Moss or Victoria British as they seem to be supplying points that have an inferior grade of alloy in the contacts. All tunstun contacts will develope a film over time, but the ones from Moss and VB seem to be worse than the ones from Burlen fuel (which are quite expensive). You may be lucky and have your modified pump run for many years trouble free (several that I did that way for clients have run troublefree for nearly 20 years, others didn't last for 500 miles). For more information regarding SU fuel pumps, see my article at: http://www.custompistols.com/cars/articles/dd_su_fuel_pumps_101.htm
Cheers - Dave
David DuBois

This thread was discussed between 12/01/2004 and 18/01/2004

MG MGB Technical index

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