Terrible cutting out/bogging down at 3500 - RESOLVED

[Andy348]

Engine is an M20B25 (1987) into a 325e (1986).

I've had this problem ever since I got the new engine into the car. While it was out of the car I changed the timing belt, water pump, oil pan and valve cover gasket. I also changed both the in-tank and external fuel pumps for used ones (apparently working) because of a high pitched whine from the external pump.

Problem:
The car starts up and runs pretty rough. When I drive it, I can't get past certain RPMs in certain gears. Even if I'm flooring it, it won't go anywhere. Quite often I can wait a second, clutch in and rev it past where it limited me before. It seems to be a problem mainly under torque, as it is less evident with the clutch in, and is more evident while going up hill.
I've currently got a fuel pressure gauge T'ed up at the rail, and this is what I've found:
Cold:
1st gear -> Cuts out at 3500rpm -> fuel pressure begins to drop slowly from 41psi while continually pressing throttle
2nd gear -> Cuts out at 3200rpm -> fuel pressure is down to 25psi
At this point, if I clutch in and wait a second or two I can rev it to 5000+
Off start up, I can rev in neutral to nearly redline.
After driving for awhile, 1st gear started cutting out at 3000rpm and 2nd gear at around 2800rpm.

So far I've replaced the CPS, sanded down and thoroughly checked the cap and rotor (it looks practically brand new), checked plugs, tested TPS and other things I'm sure I've forgotten.

I have disconnected the battery for 5 minutes and then driven it with little to no difference. I got to approx 4000rpm in first gear where it cut out.

It had pretty old gas in it, so I filled it with new gas which helped it run slightly smoother. I'm afraid there's still some bad gas in the tank, but I assume that it still has some old gas.

Also, there are a couple of unplugged wires in the glovebox. I've read that they're for the tachometer, but I'm not sure if they could be a cause of any of these problems.

Needless to say, I have no idea what the problem is and I highly appreciate any input.

Note: I am by no means an expert mechanic, so it could very well have been something I did wrong throughout the swap.

Thanks.

[jlevie]

The cause of the bogging is the loss of fuel pressure. Other things could be going on, but the fuel pressure problem must be addressed first.

Connect a test light to the fuel pump power lead and repeat the pressure test. If the light dims or goes out when you see the drop in fuel pressure the cause is electrical. A bad fuel pump relay, bad main relay, loss of timing data, bad engine harness, bad DME, or a DME power or ground problem are all possibilities. Further testing can eliminate some of those and parts swaps can eliminate the rest.

If the light doesn't dim or go out, the soft lines, filter, pumps, or FPR are suspects. If the tank is less than full, a failed or weak transfer pump will starve the high pressure pump. In a like manner a tank with a lot of rust in it will result in the transfer pump's screen clogging and starving the high pressure pump. Obstructions in the feed side of the fuel system will reduce flow and thus rail pressure . Or the high pressure pump could be weak and not able to provide enough flow to maintain rail pressure as flow through the injectors increases.

[DaButt]

Sounds like maybe you're running rich? Anyway, I very similar problems when I swapped my DME.

Bought a new (used) DME from German Auto Parts (ebay seller) and it worked like a charm.

Could be your problem, but I'm not a doctor.

[Andy348]

Okay, update.

My mechanic disconnected the in-tank fuel pump and ran it solely on the in-line fuel pump. I hit 6K in first gear and second before I needed to stop due to lack of road. I am planning on driving the car back (45 mins) on Wednesday.

Now I've read that the car can run with only one fuel pump, but it causes premature failure. Therefore I'd like to fix this problem and run it with two fuel pumps, but I'd like to avoid this same problem again.

So do I buy a new in-tank fuel pump?

edit: here are pics of my current in-tank pump and a new one I stupidly purchased:

[jlevie]

The pump you purchased is for a late model 325 with the 63L tank. You need the early style transfer pump for your car. You can't use that pump with the early 55L tank.

Your car, with the two pump system, will work okay with a failed in-tank transfer pump if the tank is full. But once the tank drops below 3/4 full the high pressure pump will start to starve.

[Andy348]

The pump you purchased is for a late model 325 with the 63L tank. You need the early style transfer pump for your car. You can't use that pump with the early 55L tank.

Your car, with the two pump system, will work okay with a failed in-tank transfer pump if the tank is full. But once the tank drops below 3/4 full the high pressure pump will start to starve.

So you recommend buying a new in-tank pump?

I'm just confused as to what could be wrong with the current in-tank pump. I mean, its obviously the problem, but what could be wrong with it? Only thing I can think of is its pushing too much fuel to the engine which is causing my problems, but I'm clearly no expert lol.

[jlevie]

The transfer pump isn't pushing too much fuel to the engine, but it may have failed and isn't supplying the high pressure pump with sufficient fuel. You can find out if the transfer pump is working by disconnecting the output line, connecting a hose to that fitting and leading the hose into a container. Then jumper the fuel pump relay socket to get the pump to run. The spec for the flow rate of the transfer pump is in the Bentley.

The other possibility, which I mentioned in my original post, is that the tank is full of rust. That rust will collect on the inlet of the transfer pump and choke off flow. Once you remove the pump you can use a stick to stir the bottom of the tank to check for a layer of rust sediment. It is also possible, especially if there is rust in the tank, that the filter could be clogged or the high pressure pump is worn out. Your fuel pressure checks clearly showed that something is taking away fuel pressure.

[Andy348]

I'll check the rust and test the in-tank pump.

My only issue is that if it runs without the in-tank pump electrically connected, the car runs fine. This leads me to believe that even if the in-tank pump had failed, I wouldn't be experiencing this in the first place, because its 'off' either way - broken or not broken. The fuel filter is new, and I don't think the high pressure pump could be the problem because it can support the engine on its own.

Excuse my naivete, but if there was rust in the tank, wouldn't that impede the flow through the in-tank pump? I'm assuming that even with the in-tank pump disconnected, the fuel is still running through it. If it was blocked with rust, wouldn't I be experiencing the problem without the in-tank fuel pump on?

Sorry to be questioning you, and I really appreciate your help. Its just I'm trying my best not to throw even more money (that I don't have!) at this problem.

I'll do the tests Wednesday.

[Andy348]

Stupid question: where does the hose from the filter go? Into the regulator or into the fuel rail close to the firewall?

edit: nevermind, not my problem...

[jlevie]

I'll check the rust and test the in-tank pump.

My only issue is that if it runs without the in-tank pump electrically connected, the car runs fine. This leads me to believe that even if the in-tank pump had failed, I wouldn't be experiencing this in the first place, because its 'off' either way - broken or not broken. The fuel filter is new, and I don't think the high pressure pump could be the problem because it can support the engine on its own.

Excuse my naivete, but if there was rust in the tank, wouldn't that impede the flow through the in-tank pump? I'm assuming that even with the in-tank pump disconnected, the fuel is still running through it. If it was blocked with rust, wouldn't I be experiencing the problem without the in-tank fuel pump on?

Theoretically, there shouldn't be much if any difference in operation of the fuel system if the transfer pump is disconnected, providing the tank is full. Though the high pressure pump will be slightly starved for fuel. But, if there is rust in the tank there can be a difference. More suction is generated at the transfer pump inlet screen when that pump is running. As the car is being driven rust in the tank get stirred up and collects on the inlet screen, eventually choking off flow. There is less suction on the inlet screen when the pump is disconnected because the high pressure pump is slightly starved because it is having to lift fuel.

If rust is a factor tests with the car stationary may not show it. The rust will just lay on the bottom of the tank. The sloshing that occurs while the car is being driven stirs up the rust and it collects on the inlet screen. When the engine is shut off that rust falls off and settles out.

[Andy348]

Theoretically, there shouldn't be much if any difference in operation of the fuel system if the transfer pump is disconnected, providing the tank is full. Though the high pressure pump will be slightly starved for fuel. But, if there is rust in the tank there can be a difference. More suction is generated at the transfer pump inlet screen when that pump is running. As the car is being driven rust in the tank get stirred up and collects on the inlet screen, eventually choking off flow. There is less suction on the inlet screen when the pump is disconnected because the high pressure pump is slightly starved because it is having to lift fuel.

If rust is a factor tests with the car stationary may not show it. The rust will just lay on the bottom of the tank. The sloshing that occurs while the car is being driven stirs up the rust and it collects on the inlet screen. When the engine is shut off that rust falls off and settles out.

Okay, that makes sense. I'll first reconnect the in-tank pump and have it spew fuel into a bucket, then I'll check whether its rusty or not. Next, I'll remove the pump and check the rust level with a stick.

[dnguyen1963]

Tie a magnet to a stick then swirl it around inside the tank. The magnet will pick up a lot of the rust particles. Make sure your magnet is secure. You do not want the magnet to get stuck to your gas tank

[Zingabe30]

I'm having a very similar problem which has now progressed to no start... but sounds exactly like what your experiencing now.. sorry. . i know how bad it sucks to be driving and then glug glug going nowhere fast

[Andy348]

Update #3: Drove the car home today (45 mins). First, second, and usually even third is fine up until redline, but 4th and 5th get cut out hard. If I'm on the highway in 4th, doing 100 at 2000rpm or so and I floor it, it may speed up to about 130ish but then completely hit a brick wall. Sometimes it'll start slowing down, other times I can just keep my foot planted and it won't go past 3500ish. I tried downshifting to third and giving it more gas and still nothing.

But first and second work flawlessly.

Does this change any diagnoses? I'm going to check for rust and such in a couple of mins.

[Andy348]

Connected the fuel pressure gauge and drove around. Seems like any time you give it some throttle the fuel pressure goes down to zero quite quickly. When I clutch in, the fuel pressure rises again.

Rear pump on/off doesn't make a difference.

I seem to be able to drive it as long as I stay under about 3000rpm and granny shift.

edit: disconnected the battery and let it sit for 15 mins. Reconnected and drove it; idle pressure seems to be down at around 30, any use of throttle makes that figure go much lower. Giving much throttle in first gear results in the pressure going down to zero.