Fuel pumps

[DeathBus]

hey guys thanks, I think I am going to use a bus pump.

[raygreenwood]

Vwbill, changing electric motors on the pumps is not possible. It is not a seperate unit. The motor is liquid cooled by gasoline flowing through it from a port inside of the pump head itself. There is even some conjecture...that the pumps may be having problems...because they were originally designed to running on leaded fuel. These are brushless motors. The fuel provides cooling and lubrication. Also, them motor rarely is the first part to go. Chances are if the motor does not turn, the pump rollers are shot too. Ray

[MGVWfan]

For an image of the innards of a D-Jet pump, got to the below link...

http://www.tiserves.com/VW/downloads/pd ... -Type3.pdf

Page 5 has a cutaway view of the pump, and a description of how it works. It's a PDF file, so I can't paste a copy of the image in this post, sorry!

BTW, as I suspected, it is a brush-type pump with a commutator. As long as the commutator is submerged, there's no oxidizer, and no sparks, so no "boom"!

Ray, you found a pump that's 1 1/8" dia Wow, that would fit in through the fuel feedline bung on the bottom of a T4 tank. Good info!

[ubercrap]

Awesome link!

[raygreenwood]

Wow....I thought they were all brushless. Some of them are.
Yep, the feeder pump can also be attached to he bottom of the fuel level gauge unit and put in that way. It has a little strainer on the bottom. Ray

[MGVWfan]

A dissertation on DC motor design...

Brushless technology requires solid-state electronics capable of switching inductive loads at 12V and 2-3 A (and temperatures usually seen in automotive applications), and transistors with the required ratings available for cheap mass production weren't around until the early 70's, and weren't commonplace until the late 70's...and it appears the Bosch D-Jet pump design dates from the 60's. A brushless design also requires some sort of position sensor, which can be a troublesome trigger coil, or a Hall-effect sensor. It was really the Hall-effect sensor that made brushless motors possible in mass produced applications. So, from the POV of the Bosch FI designers in the late 60's, a submerged commutator design must have made sense.

Others got around the problem by using a transistor oscillator pulsing a solenoid connected to a pumping diagphram (like the Hitachi designs used on various carburetted Japanese cars up through the 80's)...no need to know the pole piece position, just turn the juice on and off periodically. The Faget pumps use the same technology, sort of a Bendix aviation pump (or an SU) with no breaker points. Neat, but not capable of the pressures and flowrates needed for FI.

As far as I know, brush-type DC motors are still used in FI fuel pump design. I tore apart the Carter pump from the family '94 Voyager (3.3L V6, MPFI, Bosch and MoPar design, pump lasted 140K miles), and it was a submerged commutator design. The replacement sings just like the original, so I suspect it's the same. Some enlightened OEMs are probably using brushless motors, but there's no way to know unless you tear into one.

The brushes must obviously be compatible with the fuel used, and I suspect you're right, some of the D-Jet pump woes may be due to fuel composition changes over the years affecting the brushes, and therefore the motor's efficiency (and maybe also affecting the pump components, too). Sort of like the old "Type F" and "Dexron" thing with ATF, both are ATF's, but one can't be used in place of the other due to fluid's friction characteristics being different.

[raygreenwood]

Ah....nice to know. Thats ot what I was refering to as brushless design though....and obviously I am getting my terminology wrong. The brushes in tehse pumps are not exactly like standard brushes. They were not a standard carbon material on the the ones I have disected. The way they contact the armature is a little different.
Its nice to know what you just explained. In other words...to be truely brushless they would have to have the equivalent of a shaft encoder...or a switch system to oscillate from pole to pole because there is no stationary brush to give permanent +/- pole position between shaft and armature. I should have known that...but did not. Must have been one of the classes I slept through Ray

[vwbill]

Wow some more great info! I didnt know the motor area was liquid cooled by the fuel, is that what you guys are saying or is just the pump part cooled by the gas? I remember those little cylinder like rollers in the top so are those what wears out or the casing? It seems so weird to think of fuel running inside a motor so what stops it from being a fire hazard not being in a vapor state? bill

[MGVWfan]

Yep, the entire motor and pump is cooled and lubricated by fuel, and as long as there's no oxygen in the pump, there's no possibility of explosion.

[vwbill]

Wow, just downloaded that link and its a great source of info, thx again!
I can see how the change in fuels would affect the fuel pump , it wasnt designed that way. Do you think a lead additive would help preserve the pump or maybe some kinda additive or maybe a spray on coating for the internal parts or is that just a waste of time? Hey guys would that t3 fuel accummulator help on a t4? bill
p.s. so you couldnt rebuild the bosch pump brushes since they arent common like going to a vacuum dealer for carbon types?

[MGVWfan]

Having looked at the pump on my car, and thought about it a bit, I don't think it's practical to rebuild the pump with the materials and tools we've got available to us. The real problem in rebuilding one of these pumps is getting the motor section out in one piece, and reassembling the whole thing, I'd think. The tolerances are close, so if you reassemble it with a makeshift clamp or something to hold the motor section together, I'd bet it'd either leak or not pump, or both.

I'd bet the brush material is really unique. It would have to be much harder than normal, and the binder used to hold the conductive material (carbon, copper, whatever it is) together would have to be compatible with gasoline, alcohols, etc.

BTW, the little disc-shaped fuel pulsation dampener came on most T4's, at least the ones I've seen (a grand total of 4 to date), and it goes in the pump discharge line to smooth out the roller pump's minute pressure pulsations.

[DeathBus]

Having looked at the pump on my car, and thought about it a bit, I don't think it's practical to rebuild the pump with the materials and tools we've got available to us. The real problem in rebuilding one of these pumps is getting the motor section out in one piece, and reassembling the whole thing, I'd think. The tolerances are close, so if you reassemble it with a makeshift clamp or something to hold the motor section together, I'd bet it'd either leak or not pump, or both.

I'd bet the brush material is really unique. It would have to be much harder than normal, and the binder used to hold the conductive material (carbon, copper, whatever it is) together would have to be compatible with gasoline, alcohols, etc.

BTW, the little disc-shaped fuel pulsation dampener came on most T4's, at least the ones I've seen (a grand total of 4 to date), and it goes in the pump discharge line to smooth out the roller pump's minute pressure pulsations.

Rays Idea seams the best, use a pump from another car that has close to the same PSI, this bus pump I have while rather long will still fit in the stock location, all I have to do is cut off the stock connector from the car, remember which wire is + and - and I have a relativly inexpensive fix to the problem. No rebuilding needed, unless you are a die hard OEM type/

[MGVWfan]

Yeah, agreed, finding a pump with the same flow rate and pressure that's cheap and easily obtainable is the best option. I'm thinking about that little in-tank pump he noted in case the original pump doesn't work. I did bench test it, but you never know what'll happen for sure until you test it on the car.

[Longbeach412]

someone from the T3 list, uses a Ford pump as a replacement.

[MGVWfan]

You said the naughty F-word

Personally, I don't care from whence cometh my parts, as long as they are as good as OE, or better...so if the FoMoCo pump works, and is durable, and cheap, why not?

BTW, I had two English Fords (Anglias) as my first car, and drove a '71 Pinto for a while. Sort of turned me from Fords, thought the little Anglias never broke that I couldn't fix them to get me home.