When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
The spark knock counts in my car recently got much worse, to the point where I couldn't even maintain speed on a hill. This happened fairly suddenly, however I have had a "clacking" under heavy load for some time now. It sounds a lot like a loose lifter that is clacking intermittently. I put a fuel pressure gauge on it and drove it around and to my surprise the fuel pressure is fluctuating wildly when it's getting the knock counts under load. Like switching between 50 psi and 30 psi rapidly. This of course throws off the mix by quite a bit and is causing lean knock. What should I check first for a problem like this? It seems the fuel delivery would be suspect but I can't rule out FPR either. The pump is a walbro 255 and seems to have no problem building pressure under most circumstances. After priming the pressure does drain down, but clamping the return line (smaller tube coming from the engine) makes it hold steady. Datalog shows steady battery voltage at the fuel pump. When installed the pulsator thingy was removed and replaced with a piece of fuel line. I believe the kit came with the usual "ribbed" style of hose clamps and not the smooth fuel line clamps I've seen.
Is there any good way to isolate which component might be the problem? Would hate to have to drop the tank and put in a new pump if I don't have to; although there's a fuel pump access hack from a PO, I'm not about to cut the lines on my $400 AC Delco sending unit.
If clamping the return line keeps it from dropping then the regulator isn't holding rail pressure. Start there. Get a new regulator diaphragm and optionally an adjustable regulator cap just to be prepared when you get it apart. You will need intake runner gaskets. If you at all suspect your injectors I highly recommend a set of SouthBay injectors while you are in there. If you do any tuning it's also a good time to remove the EGR and anything else you don't want.
I actually already have both an adjustable regulator and the SouthBay fuel injectors. I did install a new diaphragm when I put the regulator on. For some reason it is not fully sealing. I don't know if it's the diaphragm or the base it installs into that has the problem. Not being a human micrometer it would be very hard to tell if one of the seats is not perfectly flat, although I suppose an off-center diaphragm could do it. Certainly worth trying to either go back to the old regulator or at least pop it open.
Well you might have to figure out a way to polish the seat in the fuel rail. Something extremely flat like a small round of glass with some high 2000, 2500, and 3000 grit wet dry paper held by contact cement..... Lube with WD40.
I have not tried this. It's just what I would probably be prepared to do given your situation. Note that I have spare fuel rails if that junk goes sideways.... And I have real world experience with polishing disc valves on industrial compressors. Don't do this if you can't afford to throw the whole shootin match in the art bin.
So there's a few situations that might lead to my problem:
There's an issue with the fuel pressure regulator and it's bleeding off pressure erratically. There's definitely a minor issue with it since it bleeds off fairly quickly, so it would be good to address this anyway.
There is a restriction in the feed line or filter that doesn't allow sufficient flow. This might be somewhat difficult to test, maybe I could disconnect the feed line at the engine and run it into a tank to see how strong the flow is?
There is a problem with the pump and it's going out. Not really a good way to test this one without swapping pumps?
There's a problem with the pump not getting stable power. I could potentially monitor voltage or current close to the pump and see what readings I get.
There's a clogged fuel sock. Would require inspection; #2 check might rule this out.
The connection between the pump and the hard line has deteriorated/is leaking. This may be testable by clamping the return line, then dead-heading the pump briefly. If the pressure rises to a high limit and does not leak down then the connection should be OK since it's after the internal check valve in the pump.
Try partially clamping the return line while it's running. A couple quarters and some vice grips.... if this stabilizes the pressure then the problem is the regulator.
I doubt it's that Walbro. And a partially blocked line seems unlikely as well. What did the tank look like when you put the pump in?
Tank wasn't too bad, not super corroded or anything. I did some tests today. First I clamped the return line and ran the pump briefly up to 70 psi. It held without dropping so the fuel pump check valve and the rubber line connecting to the sending unit are fine. I then check the amps while the fuel pump was running and it read ~6.4 amps which is a good amount and indicates a healthy pump, at least electrically. I measured the resistance of the pump power wire to ground at the relay and it was 0 ohms, so there isn't any resistance in the wires or the pump commutators, so that is also good. With that information I'm going to focus on getting the regulator sorted before doing anything further with the pump side.
I also started the regulator work. I bought a fuel rail locally, and although it had some corrosion in the pipes I was able to use the fuel regulator block. Comparing it to mine, I could see a difference. When inspecting the seat with a flashlight, you could see a distinct area on my old one where it reflected at a slightly different angle. This may have been caused by the little "coin" in my adjustable FPR being way off center, potentially causing uneven spring pressure. The one I bought shimmers at the same angle, indicating that it's flat. My diaphragm looks alright as far as not being torn. I inspected the disc on it, and it had some spots where the worn-in ring did not look uniform as well as visible machining marks. I took a piece of 2000 grit and a glass table and lapped the surface of it to a near-mirror finish. As I was lapping it, I could see some parts of the ring disappear before others, and there were some deep machine marks which didn't go away until a good amount of lapping after the worn-in ring disappeared. I'm pretty confident it should seal now. Tomorrow I'll be trying to get everything back together, which will be a pain because I foolishly used a thin layer of RTV on some of the intake gaskets when first assembling. So I'll be spending a lot of time with a scraper.
Oh, and I noticed there where two different connecting tubes between the regulator block and the passenger rail. One had a huge diameter hole through it, and the other had a smaller diameter hole through it. I assume this was done to force more of the fuel to the other rail? I ended up putting the smaller diameter hole one in, so if this isn't the right one to use, let me know. I'd much rather find out before I put everything back together .
So an update: Got it all back together. Fuel pressure will sometimes hold sometimes not once the key is off. It's running a lot better at WOT, the fuel pressure stays pretty steady. There's still some problem areas like mid throttle on a hill where it still likes to jump around. Going to give it a little time and see if it wants to settle down, otherwise I'll probably have to try a different diaphragm.
So I've given it a chance to wear in, but it's still wildly fluctuating in various circumstances. It did hold 30 PSI overnight however, so when it does decide to seal it seals well after doing the sanding. Called up SouthBay and they suggested it may either be vacuum or fuel pump related. Pulled the vacuum line and drove around and it's still fluctuating a lot. I think I'll try bypassing the fuel pump relay and driving around since all my testing was without the relay in the equation. Always possible it's going out and not passing as much current as my bypass switch was. Are there any ways that I can test the fuel pump without actually dropping the tank and swapping in a new one? If I can't narrow it down to one or the other I'll just try replacing the diaphragm since it's relatively cheap and easy to access.
Pulsator was removed and replaced with a piece of rubber hose, as is typical when switching over to a turbine style pump like the Walbro. I don't think it's a flow problem since it now smooths out a lot under WOT. I suppose my cheapy fuel pressure gauge could always be giving me faulty reading but I doubt it.
It's mostly stable at idle, maybe a couple PSI either way. Under certain loads it starts fluctuating rapidly, even with the vacuum line removed. The pressure seemed to rise and fall as would be expected when the vacuum line was connected, apart from the times when it would rapidly fluctuate.
Didn't seem to be. Pressure doesn't drop under WOT, the fuel filter has less than 20K miles on it, and I didn't see any kinks or anything on my last inspection. The return line also seems pretty clear. We blew some compressed air through the rail to drain it out before popping it off and it was blowing out the uncapped tank with a strong flow of air.
Well, supposedly, they are gerotor pumps, so, if there is pressure pulses generated by the pump, they should be far to close together to be shown on a gauge.....
About the only other thing I could think of, would be if one or more of your injectors flowed a LOT more fuel than it should? Maybe do a flow test on 'em?
The difficulty in that is that it only is really reproducible under certain driving conditions, it's much harder to do at a standstill. And I don't know about you but having a bucket of fuel around while you're driving doesn't sound too safe . I think you're right though, it would be best if I can rule out as much as possible. I've got an injector tester coming in the mail. Although I don't think there's a problem with them, at this point I'm trying not to make any assumptions. I'll also try a different pressure gauge at some point.
Some additional testing I've done since the last post:
- Used a pulse tester on the injectors, they all ended up within a PSI of each other so we're good there.
- Did a volume test off the return line after the regulator, engine off. Got just under a pint in 10 seconds which is plenty of flow.
- Checked the waveform for the current on the pump. Not perfectly flat but not horrible. I'll include it below. I still need to check it while the pressure is fluctuating but I have to wait for my scope battery to come in the mail. Doing the math, the pump is running at about 4500 RPM.
It seems like its either a gauge problem or a weird regulator issue. Everything I've done so far seems like the pump is working fine, and everything south of the regulator is clear.
EDIT: To clarify on what I said on the waveform, the bumps should all be at the same level ideally, that's what I mean by flat. The bumps are expected since it's drawing current at each pass of the commutators.
I really do need to get one of those! I was actually looking at some, but was having trouble finding the necessary adapters in order to hook it up. Any suggestions on that front?
It's just 1/8" NPT. That's the same as what's in the block. Just go to a hose and fitting supplier and get some 3AN braided brake line and 1/8" NPT adaptors.
Ah, gotcha. Was wondering where there was a NPT port in the fuel system . I think with that particular adapter it would get real close to interfering with the plenum, potentially even with a right angle adapter on the top. Looks like it's -4AN though, so I can probably find something. Not too concerned about removing the schrader valve temporarily.
You could also tap into the feed pressure at the soft line connection from the engine to frame rail area. There are adaptors sold for this on summit, etc.
You could cut the hose also and put a T in the line. Done right it wouldn't be ugly but it would require some ear clamps and careful selection of fittings to not look like an abortion.
GD
Last edited by GeneralDisorder; 08-02-2017 at 02:41 PM.
I don't plan on running this all the time, so reversible is best. Right at the test port would give me the best indication of fuel rail pressure I would think. These AN fittings are so expensive though! $24 for a 4AN female->4AN female right angle fitting?! I mean it's blue and pretty and all, but geeze. Anyone got a source for fittings that don't break the bank so much?
I was thinking permanent install and powered off the car. You can replace the scope with a process control display and just have fuel pressure in the cockpit.
Certainly could do a permanent install eventually. I've toyed with the idea of having a switch which would turn the oil pressure gauge into a fuel pressure gauge and the coolant temp gauge into a transmission temp gauge, just for sanity checking. Not particularly accurate gauges though . Wouldn't mind getting it integrated into the 730 either if it has a spare input, but that's a goal for later.
07-19-2017, 07:47 PM
. 
. Wouldn't mind getting it integrated into the 730 either if it has a spare input, but that's a goal for later.