[dr.electron]

Yeah - I've encountered some incorrect diagrams. I'm confident that this particular one is correct based on looking at the controller and harness locations in my specific car.

A little more reading tells me that people who upgrade their fuel pumps to the TTRS fuel pump sometimes have problems with their stock pump controller. The upgrade is apparently common in modded TTs and other MQB platform cars, and often comes along with an upgrade to the controller as well. But....why does Audi itself tell me that smaller controller will work in the TTRS? Maybe they've improved it?

 

[TT'sRevenge]

I think there may have been some confusion when I was talking about the "wires" to the pump... I didn't mean literally there were only going to be two or three wires, heh. I meant how the pump itself is powered, forgetting about the level sensor, etc. stuff. Indeed the entire assembly has a connector with what should be 5 pins, but some are for other things--like the level sensor.

I believe the two outside pins are the ones that power the pump but should def. look at diagrams to be sure. And again it may be different on RS models, how the pump is operated, I'm not sure.

As for the "VAG way" of diagnosing the fuel pump I'm sure you've seen already seen that it requires measuring the volume of fuel by disconnecting the line and capturing the fuel that's pumped out. Indeed it's a complicated affair but it's hard to believe you had an old defective pump and then somehow got a new one...that's still defective in basically the same way. But then again the sensor we could say the same--if the old one was defective, and the new one...is also defective... Well I mean what are the chances unless we're talking about eBay/Amazon parts which it seems we are not. But stuff like this does happen from time to time.

As for the controller, from the looks of it, 8J0906071A looks completely different from the controllers typically found on say the 2.0T cars. Those ones look more like the 3C part you've referenced. So from that I would say no that part won't work and the RS uses a different controller and probably has a differently operating pump as a result. I was not really in tune to this before but yeah that controller for the RS looks pretty darn expensive. And, getting one out of a used vehicle...in the US...is like finding a needle in a haystack. If a new one is $800 and the part is rare, wreckers even if they have one will be charging $400-500 anyway.

As for diagnosing the controller via the VAG method...it seems it's a "too bad so sad" problem because it's not written out how to do that. It's "hidden" within GFF...like a lot of things. So unless you have access to a 5054A or 6154 and ODIS, you are not going to be able to entertain this avenue.

So this is not looking to great on the controller front, it's certainly not something you want to just "try to replace and see", at that price. Should probably exhaust all other avenues before that, including doing this fuel volume measurement business.

 

[TT'sRevenge]

A little more reading tells me that people who upgrade their fuel pumps to the TTRS fuel pump sometimes have problems with their stock pump controller. The upgrade is apparently common in modded TTs and other MQB platform cars, and often comes along with an upgrade to the controller as well. But....why does Audi itself tell me that smaller controller will work in the TTRS? Maybe they've improved it?

I don't think that other controller will work but what are these, presumably aftermarket (?), controller upgrades like, in terms of cost? Maybe better than the stock one?

On the controller compatibility--someone may look up in ETKA and see what is spit out, maybe that will help. But otherwise if still no good answers I suggest trying to PM a user named Smac770 on Audizine (he's also on a few other forums) and see if he can help you out with the part information. He's an incredibly knowledgeable guy who is very helpful! He seems to have a heck of a lot of information on parts as well as procedures. I'm not sure if he's a dealer tech or perhaps used to be one or what, but definitely seems to provide a crazy amount of info I've not seen elsewhere...at least for the B8 Audis where he seems to frequent the most in terms of subforums.

 

[dr.electron]

@TT'sRevenge - thanks for the ongoing participation here. It definitely helps to have thoughts from someone else's head to throw in with mine. I knew what you meant regarding wiring - I just made the mistake in my interpretation of how the pump and controller are connected together.

My plan for later today is to go and look at the voltage actually applied to the pump by the controller (as opposed to looking at the duty signal from the ECU - which was pointless, as it turns out) - and I'm expecting that it will be lower than it should be. I agree that the probability of replacing what was clearly a bad old pump with a bad new pump is extremely low, as is swapping in a bad new sensor. The controller is the obvious suspect here.

I agree that it seems unlikely the 3XX part will work. And if it does work at first, I'd be concerned that it would eventually overheat and stop working. The only way I'd use that part is if I can find definitive language from Audi stating that it's up to the task. But I don't think I'll find that. I'll see if I can find Smac770.

The aftermarket pump controller I find first is the Torquebyte PM4, which costs about the same as the OEM controller. It's intended for use with TTRS pumps installed in modded cars, where the stock wiring harness isn't sufficient to carry the power for the TTRS pump. It goes in the spare tire bay and gets battery power directly from the battery. That controller is the same price as an OEM one and would be a lot easier to install, since getting the OEM one out appears to require dropping the gas tank a couple of cm. I could probably rig that up, but it would be a pain. For the same price, I may as well get a better controller that's easier for me to install.

 

[TT'sRevenge]

Oh so the RS controller is not even mounted like the "standard" controller either--if you have to drop the tank to get it out, that's def. different. The other controllers (like the 3C) one are mounted on the "inside" and are easily accessed.

 

[dr.electron]

Yeah - it's mounted down under the car next to the tank. It's a good bit bigger than the 3C controllers. Though, honestly, I could potentially stick it somewhere else. The only concern would be making sure it had a place to dump heat. It's mounted to a metal bracket that's attached to the frame of the car right now. But there's other metal around...

 

[Vassilis]

Yes - it is, now that you mention it. The previous owner put a United Motorsport tune on it. I have the boost set to the minimum (you can set it with VCDS) - which is 20psi. But I've been driving the car for a year, and this is the first time I've had fuel issues.

Bad remaps can typically cause pressure errors which cannot be solved even with a new LPFP and HPFP, I have seen a few on this forum during the past years. I had a similar problem when I upgraded to the Loba HPFP created by the tuning software, which was easily solved by the tuner with an new map.

 

[dr.electron]

Bad remaps can typically cause pressure errors which cannot be solved even with a new LPFP and HPFP, I have seen a few on this forum during the past years.

The only reason I'm not quick to blame the remap is that the car has been running with no issues for the entire year and several thousand miles while I've owned it, and I bought it with the tune already installed.

 

[dr.electron]

Tonight I measured the voltage applied to the pump by the controller while the car was idling. What I expected to see was a rise in duty cycle at some point - as I've seen in the past - but a disproportionately small rise in applied voltage. This would tell me that the controller was not delivering enough power. To get to the pump drive pins, I had to hack together some clip leads and contacts:

Yes - I fully expected this to be a poor connection as compared to having the connector directly attached to the pump. But I was looking for a fall off in voltage disproportionate to duty cycle.

But what I found was that it took a much longer time for this setup to rail duty cycle. And even then, the pressure stayed higher than before once it railed. Yesterday it took about 12min to rail duty cycle. With the mess you see in the picture above, it took 25min. And by the time I stopped the test, the clip leads were very warm, as you might expect.

I stopped the engine, and duty dropped back to 9.8% and held 4.5bar or so. I unhooked everything, plugged the controller back in directly, and turned the ignition back on - 9.8% and 3.3bar. I started the engine - railed to 68.7% and gave me about 2.6bar.

I put the clip leads back in place, and with the ignition on I was back to 3.7bar at 9.8%. Started the car and got a little under 3bar.

So overall, my clip lead hack performed better than the direct connection. Which I did not expect.

All of this suggests that the controller's inability to drive the pump is potentially due to the quality of the connection between the controller and pump. That connection has resistance, and as it warms up the resistance will increase. Eventually the connection is poor enough that the duty rails as the pump tries to maintain pressure.

So how to improve the connection? Re-pin the connector? Will that fix it? I may also have poor connectivity in the other connector on top of the fuel pump, too. In general, it looks like I need to reduce impedance between the battery and the controller and between the controller and the pump. If I can get the pins out of that connector, I could at least look at them and see if they're questionable.

The most robust solution to this is an aftermarket pump controller - like the Torquebyte PM4 - that will connect directly to the battery in the trunk and then to the pump. This is probably overkill - this controller is designed to drive more than one pump at a time. But If that doesn't do it, the pump has to be a problem.

 

[Jezzie]

I would at least clean all the connector pins with a decent quality contact cleaner like De-Oxit Gold which is what I use.

 

[dr.electron]

Cleaning the contacts is a good suggestion - particularly the contacts in the connector, which I will pop out of there to inspect. I have a couple of other tests to run that should tell me definitively if the pump is performing as it should, which I'll undertake this weekend.

Does anybody know how much current that pump draws at idle? I know this is a long-shot question to ask, and I'll answer myself this weekend.

 

[dr.electron]

Good news today. I spent some time ensuring the best possible contact between the fuel pump controller connector and the pump itself. I used some small tools to press the contact surfaces in the controller connector terminals together a bit to improve clamp force on the pump connector surfaces, and I cleaned both the controller connector terminals and the pump connector terminals with a product similar to what @Jezzie recommended (it was just a different brand of terminal cleaner). I plugged everything in firmly, and went for a drive. Here's the data:

To start with, the idle duty cycle was the lowest I've ever seen it since I started this investigation. It's typically run at 41.6% to 42.5% when things were "working", but this time it sat between 38.4% and 40%. After 18 minutes of driving - including several hard pulls to push the fuel system - idle remained at a solid 38.4% - 39.2% for another 2 minutes of sitting in my driveway. The controller has typically railed duty cycle well before this, even at idle (see my post from Oct 6th with the plots).

Based on the performance of the low pressure fuel system today, I'm concluding that my most recent issues were related to the quality of electrical connections to the low pressure pump. As a summary, here's the sequence of events for my fuel system adventure over the past few weeks:

1. I got a CEL for low pressure fuel regulation, pressure outside of specification
2. I responded by doing the "easy" things and replacing both the fuel filter and the low fuel pressure sensor on the fuel rail at the engine. The sensor was the Audi part, and the filter was from Mann.
3. I got the same CEL again, checked the fuel pressure while idling and found that even then the pump could only manage a couple of bar. Plus, I could hear the pump "squealing".
4. I replaced the low pressure fuel pump with a Continental pump from ECS Tuning. So I expected it to be a good component.
5. Initially, the system seemed to be good. I was running 42.5% duty at idle and holding 4.5-5bar. I drove around a little bit and the pressure never dropped below 4 even when I pushed the car. But I didn't log cycle.
6. While driving the next day, I got another CEL for low fuel pressure
7. This launched me into the detailed investigation captured in this thread