Printable Version of Topic

Posted by: brant Sep 10 2016, 07:54 AM

I think there must be a check valve built into the 3 port djet pump

I believe mine is not working
(Hard starting symptoms)

Has anyone pulled theirs apart?

Pictures or a thread?

I'm also wondering if adding an in line check valve would work?

Posted by: sixnotfour Sep 10 2016, 08:03 AM

its a pressure bypass , not a check valve

Remember the pump runs a few seconds to pressurize the system before cranking..

whats your hurry man, relax.. let it work

Posted by: brant Sep 10 2016, 08:48 AM

Ok it's just a bypass

So has anyone ever added a check valve?
I imagine it adds some restriction to the system
What would be the other reasons this is a bad idea?

I'm having a hard start situation
If I turn the key it is a 25 second crank before it starts

If I turn the key off/on about 5 times and let the pump prime
Then it will start instantly

So I'm loosing pressure
I figure it's either a bad/leaky injector or a situation where the system s draining back to the tank

I will test fuel pressure and injectors when I get home
But might also test the drain-back by adding a check valve?

Posted by: sixnotfour Sep 10 2016, 08:56 AM

filters dirty or pumps bad , use a 2port later pump..relax

Posted by: Dave_Darling Sep 10 2016, 10:49 AM

The pump itself acts as a check valve, at least to some extent. There is no specific check valve built in. There is a spring-loaded valve, but its intent is to bleed off overly high pressure from inside the pump back to the tank.

--DD

Posted by: brant Sep 12 2016, 10:22 AM

Dave,

I'm having a Djet problem where after the car has set for one minute (or even over night) it is hard to restart

cranking the starter for approximately 30 seconds works, and everything runs fine
turning the key on about 4 times, to prime the fuel system yields a totally normal quick start

(I can hear the pump run for the 1 second activation when the key is turned prior to engaging the starter)

Prior to this problem starting up, I had tested/rebuilt all of the injectors and verified fuel pressure

obviously I need to check the injectors again for a possible leak/failure and check the fuel pressure again

the only winter modifications in the induction system were new hoses and relocating the pump from the front belly pan area, to the just below the tank area (a la 1975 style)... so the pump moved about 10 inches higher into the chassis and I suppose that could effect the bleed back a little?

I was never boiling fuel or having a hard to start problem before
I verified fuel pressure after the fuel pump slight relocation

I'm wondering if a failed fuel pressure (stock style) regulator could fail and allow all of the sitting fuel pressure to bleed off down the return line too.

my current working theory is that
1) pressure is bleeding out an injector (although it doesn't seem rich when it starts)
2) pressure is bleeding out through the feed line or pump
3) pressure is bleeding out through the return line or stock style pressure regulator

I have not tested yet.
I will be able to test this weekend to verify that the pressure is bleeding off
but during the 1 second pump prime with the key on, I can hear the pump change tone as if it is re-pressurizing the system and this is what led me to the de-pressurized-system theory.

gas mileage is still 30 -34mpg on the highway so I doubt the injectors are leaking but need to verify.

thus my thoughts about a check valve.

brant

Posted by: John Sep 12 2016, 11:08 AM

My suggestion would be to recheck the lines under the tank.

If you are using a 75 style pump (inlet one end, outlet opposite end) then one of the two lines typically makes a 180 bend. My solution to having a kinked line under the tank is to take a piece of Stainless tubing and make a smooth 180 in that and use that for the supply side from the tank to the pump. I only use enough hard line to make a "U" so that I can get that out the access hole into the front trunk. With that in mind, I think I prefer them mounted just above the gravel pan next to the steering rack.

Since you are using a old 3-port pump, I would suggest that you double check to make sure you are using the correct ports to pressurize the fuel ring and return to tank.

Good luck.

Posted by: Eric_Shea Sep 12 2016, 11:12 AM

Time for one of these:



Any of the Denver crew have a fuel pump you could try?

Posted by: Elliot Cannon Sep 12 2016, 11:13 AM

I had an issue like that with a '79 911. It would crank for some time before starting. My Porsche guy said it was a check valve that maintains pressure in the system when the engine isn't running. When the pressure bleeds down it take some cranking so the pump can get the pressure back up. Don't you carry a couple spare fuel pumps in your trunk?

Posted by: brant Sep 12 2016, 11:21 AM

I can easily pull the pump this weekend.
I added a 75 style fuel door to the passenger side of the front trunk
I really don't think I have any kinked lines, but can verify easily by opening the door and shinning a flash light


oddly I've put a few thousand miles on the car since the new fuel lines were added 5/16 and this is a new problem

but I do have a brand new spare pump that I can install
(after I verify pressure and injectors)

I really do think my fuel pressure is bleeding down somewhere
I don't think a kinked line would cause that... and once running everything is fine
(400miles yesterday and it ran like a top, getting over 30mpg the whole time)


Yes....
I need to restore it
I just need to pay off one more race motor and strip it down... then its definitely time

Posted by: John Sep 12 2016, 07:30 PM

From what I recall (this is from memory) on the 3 port pumps, if one had the supply line from the tank hooked up correctly and had the other two ports swapped (pressure port from pump "Teed" into the return line to the tank) and the return port from pump feeding the fuel ring, that it might build enough back pressure through the smaller return tube to the tank that the car would eventually start and run. These stock 4-cyl engines really don't consume that much fuel, so I believe that once enough pressure built up, the car might run fine, but I believe it would take time to build enough pressure in the first place.

It can't hurt to check the ports on the pump just to make sure.

There is no check valve in the fuel circuit in a stock 4-cyl 914 that I am aware of (just the pressure relief on the 3-port pump).

Posted by: Dave_Darling Sep 12 2016, 08:35 PM

Definitely check the pressure. You're only speculating until you see what the pressure actually does.

You might also want to check that the four-pin connector on the relay board is fully seated. One of the pins there carries the starting signal to the ECU, which in part uses it to tell the fuel pump to run.

--DD

Posted by: rhodyguy Sep 13 2016, 12:51 PM

Have you looked down the fill opening and glaced at the filter sock recently?

Posted by: brant Sep 13 2016, 01:02 PM

when I had the tank out last april-ish

but it runs great
I had it to about 115mph sustained on sunday for approximately 5 miles
no fuel starvation while running

just no fuel to start for about 20-30 seconds while it primes up.

Posted by: John Sep 13 2016, 07:14 PM

Just for giggles, you might want to check pressures. Also as a test, pinch off the return line (after the pressure regulator) and see if you get a faster start.

Posted by: sixnotfour Sep 13 2016, 08:25 PM

put a 2 port pump in,, problem gone..the roller vane pump needs to be lower for feed ..total restoration not necessary at this point ..drive it like you stole it...


you are correct..like it needs a check valve ,,really you need the proper pump for where you put it..


In fact I send you one free ..I don't do type 4sale anymore....LMK

Posted by: brant Sep 13 2016, 09:02 PM

I will buy one from you if you have a spare..
I'm thinking the relocation is part of my problem

Posted by: sixnotfour Sep 13 2016, 09:15 PM

pm me address

Posted by: brant Sep 19 2016, 07:50 PM

Question

I know the injectors fire in pairs
When I pull the injectors and turn the key on I get the 1 second of pump running without turning the motor over

2 of the injectors spray for that 1 second
If I unplug their electrical feed none spray
Does this mean that the distributor has stopped on the pair that are firing?
None leak with those two electrical connections unplugged

Pressure gauge builds up for that one second then bleeds down just as quickly as it built up

Posted by: brant Sep 22 2016, 11:40 AM

So Brad Anders website indicates their is a check valve in the fuel pump and that there have been instances of them failing...

unfortunately the URL's on his page are not working:
Fuel Pump •Function: Supplies fuel to the injectors and cold start valve.
•Failure Modes ◦No pressure/inoperative: Can be due to electrical or mechanical failure. The engine will not run in this condition. See the URL below for a test procedure.
◦Improper fuel line hookup: Very odd problems in maintaining pressure, with odd drivability problems. Check the factory workshop manual for the correct hookup.
◦Clogged return line: I had this happen when I converted back from carburetors, my return line was clogged at the tank. High fuel pressure and rich, possible gas contamination of oil. If you can't regulate the pressure, this may be your problem.
◦Failed Check Valve: The check valve keeps the pressure to the injectors and CSV up after the fuel pump is shut off to prevent vapor voids from forming and speed up starting. A failed check valve may cause some hot starting problems. There has been one report of a failed check valve causing the fuel pump to return all fuel to the tank, with no supply to the injectors. See the URL below for a test procedure.

•Notes: Detailed procedures on checking the fuel pump are documented in the Factory Workshop Manual. The electrical plug, contacts, and boot are often in poor condition due to exposure to heat, battery acid, and the environment. Contact failures are a common cause of a sudden shut-off of your car while driving. You should hear the fuel pump run for 1.5 seconds after turning the key to the "on" position. If you don't hear the pump run, check the fuel pump relay on the relay board, and the connection plug to the fuel pump. I keep a cheap Radio Shack multi-meter in my car just for this problem.
•More: There is an excellent reference on debugging fuel pump problems at the URL below:


http://www.914fan.net/fuelpump.html

I also have created a series of flow charts for diagnosing the fuel pump circuit and relays:

http://members.rennlist.com/pbanders/FPChecklist.htm

Posted by: BeatNavy Sep 22 2016, 01:22 PM

Brant - I have similar experience on an otherwise generally well-running 2056 D-Jet. I've wondered the same thing about a check valve. I bought my fuel pump rebuilt from AA a couple of years ago, but I'm thinking about changing to that Bosch 2 port unit.

Can someone confirm: should the fuel pressure in the system STAY ~29 psi even when the pump is off (or maybe bleed down slowly)? Mine does not. It bleeds off right away. My other symptoms:

1. Takes me about 5 seconds of cranking to start when it's cold, and I ALWAYS have to feather it once or twice during the first 10 seconds of running or it will die one time. If it does die, it will immediately fire up with turn of key and stay started.

2. If I kill engine and IMMEDIATELY restart, it fires up with no problem.

3. If stop briefly for a few minutes, e.g., to fill up gas or run quickly into a store, it may take 5 to 15 seconds of cranking and then will idle so low for 20 or 30 seconds that I normally have to feather it to keep running. After that, idle picks up and all good. It doesn't seem like it's that "warm start over-enrichment" issue, because the head temp is still high, and my O2 is showing more like 14+ at that point.

This morning after a 20 minute doctors appointment. I started the car with accelerator depressed, and it did seem to help it start more quickly and stay started.

Maybe I'll put replacing the fuel pump on my to-do list and see if that helps.

Also, I've tried to go to that P914fan.net link before hoping to get answers, only to find a "fitness fan blog." And Brad's fuel pump page isn't working the way it's supposed to anymore (I think it may not be compatible with modern browsers). I know he's been active here recently - I don't know if he's got the time to update it or not, but it would be appreciated

Posted by: 87m491 Sep 22 2016, 01:51 PM

I don't know D-jet at all compared to my CIS cars but I think it would stand to reason that residual pressure might bleed off but not drop to zero like flipping a switch.

5 Seconds to start from cold could be almost anything, then feathering might point to another tuning issue as the whole purpose of F.I. is to preclude this issue. Most of what I have read and experienced on D Jet is a high initial idle with a drop towards normal once warm.

Killing and re-firing instantly doesn't sound too surprising with everything else, pressure may be bleeding off but every turn off the key on fires off the pump with a short prime pulse giving you back "full" pressure after the quick shut down.

Your short stop scenario is like classic CIS vapor lock or pressure bleed down via a bad check valve. Since after the initial prime the pump won't run unless it senses the engine running and of course the engine won't run without higher fuel pressure. Many will hard wire the pump to run with the ignition on circumventing the speed sensor, and causing a safety hazard.

If you can't find a source for the "pressure leaks" I'd say get the new 2 port pump and be done with it.

Posted by: brant Sep 22 2016, 04:03 PM

That sounds like exactly the same symptoms
I too would love to know if a stock djet system maintains some pressure when shut off
Mine maintains zero pressure within 1-2 seconds

I will let you know soon what my findings are from removing/testing/and perhaps beating my pump with a hammer. I intend to remove the spring loaded return over flow valve and examine it

If I don't find anything out. I intend to get a 2 port pump and test it also

Posted by: BeatNavy Sep 22 2016, 06:09 PM

Brant - that would be awesome. If somehow I get around to swapping pumps before you do I'll make sure to post an update. It's a small but annoying behavior that does sometime impact my driving -- so I want to get it fixed

Thanks!

Posted by: Dave_Darling Sep 23 2016, 09:56 AM

My system bleeds down to ~10 PSI relatively quickly, over the course of a minute or so, and then much slower down to zero. Dropping to zero within seconds tells me something is leaking.

There is no check valve inside the three-port pump. The rollers and rotor in the pump serve as a check to reverse flow, that's it. The valve in there is only a pressure bleed-off to dump overly high-pressure fuel back to the tank; it is not a one-way check-valve.

I would suspect your pressure regulator may be leaking, or possibly an injector is stuck open.

--DD

Posted by: brant Sep 23 2016, 11:47 AM

How do test the pressure regulator?
It sits at 29psi when turning over and 29psi while running

I pulled the injectors and they are not leaking

Brad Anders site confused me by claiming there is a check valve in the pump

Posted by: BeatNavy Sep 23 2016, 12:17 PM

Brant, did you test the injectors for leaking while cranking or with the system pressurized? I probably need to do the same, I found a pretty decent write up of some injector tests on this thread:

http://www.914world.com/bbs2/index.php?s=&showtopic=77265&view=findpost&p=958085

Curiously, he mentions a "check valve" in the pump too. Somehow there's a notion of a check valve, or some function that maintains pressure in the system, inside our stock pumps.

Posted by: Dave_Darling Sep 23 2016, 04:27 PM

I think the references to "check valve" either misinterpret the over-pressure valve, or are references to the fuel not being able to flow backwards through the pump stage.

To test the regulator, you can get an extra piece of fuel hose and run it from the regulator outlet to a good-sized container. Run the fuel pump until you get to 29 PSI, then stop. If a lot of fuel keeps coming out of the regulator outlet while the pressure drops, the regulator is probably letting the pressure out.

If no fuel comes out, and the pressure drops, pull the injectors off of the engine while leaving them connected to the fuel lines. Check for leaks. Ditto the cold-start valve.

--DD

Posted by: brant Sep 23 2016, 04:40 PM

Already tested the injectors but will test the regulator next

Also is it confirmed that the regulator should hold fuel pressure when working correctly?

Posted by: brant Sep 23 2016, 07:32 PM

I fixed it

Will keep testing after it cools down for a couple of days... But 2 seperate tests of 30 minutes show good results.

An open fuel pressure blow off valve seems to offer a pathway to bleed off the lines and bypass the pump rollers. Causing the hard start

Not a check valve but fails in the same way as a check valve

My valve failed. There is a piece of rubber holding it together, allowing it to work correctly...


Attached thumbnail(s)

Posted by: BeatNavy Sep 24 2016, 05:56 AM

Did you have to break open the pump to get to that pressure relief valve? And you were able to either reseat the rubber or replace it with something?

Good info! I may try to find my "spare" pump...

Posted by: brant Sep 25 2016, 09:22 AM

Some details
I may get a better picture when I get home

So on the 3 port pump there is a pressure blow off valve that opens a pathway for the pressurized fuel to bypass the output "D" fitting on the pump. I assume the original designers figure that when the hose feeding the injectors was already fully pressured to 29psi (built to that pressure by the regulator) that the pump was pushing against that head pressure and possibly running under load needlessly (wear to the pump gears?)

What ever their original logic was, they created a blow off valve for the fuel to bypass that out put path and divert to the tank return pathway

when the valve is open and doesn't get pushed by its 2springs back to the closed position, and the car/pump is turned off..... The residual fuel in the lines is sitting At 29psi and runs backwards in the circuit to drain through this open valve when sitting at rest

There is a crimp cap on the pump
Under the cap is:
a cap gasket
A large spring
A spring plate/divider
A small spring
A rubber snapped into the piston (40 years old)
And a piston

My rubber broke (a bad pregnancy joke)

The one end of the rubber creates the seal for the closed piston
With the rubber missing the piston doesn't close and leaves this alternative pathway for fuel to drain backwards in the circuit when the pump/key is turned off

You can pry the aluminum cap off of the pump and expose all of these parts

I will look for replacement part alternatives when I get back home
Out of town right now working on the racecar