SPICA anti-leakback valve

[redblk63]

My Sport sedan does not start from a several day rest in a reasonable length of time unless I prime with ether spray, or just dump some fuel into the intakes. Once started, it runs OK.

I surmise this is because the little valves in the 4 towers at the top of the Fuel injection pump (bottom of the fuel injection pipes between the pump and injectors) allow gas in the injection lines to leak back into the pump.

My question is this: Does anyone remember reading an "Alfa Owners" article about how to fashion a device to remove these valves so they can be carefully lapped?

I know that a pipe with 12 mm ID tapped with a 1.0 mm pitch thread would work, if the OD is less than 16 mm so it will fit into the tower. Trouble is, I don't have that. This article had some trick, I think some commonly available SAE size something was close enough to work.

How about it? Has anyone else had cause to jury rig a tool for this?

I can remove the tower itself, and the moving part of the valve. It is the valve body which needs to be pulled from the pump. I used needle nose pliers on my "practice pump", but don't want to mes up the good one.

Thanks,
Larry in Pasadena, back from the 10th and last (??) No Frills Iron Bottom

[Roadtrip]

Larry - Here's a quote from a Wes Ingram article:

"Delivery valve failure usually shows up at low speed. The symptom will be: cruise and acceleration are good, but has a rough idle, and emission levels are high. Adjusting the fuel mixture solenoid does not correct the problem.

If the car has been out of service, the delivery valves can rust. I do not recommend attempting delivery valve repairs in the field."

Was your Spica pump left sitting derelict recently since this problem began?

Before getting to the really obscure stuff like delivery valves, I'd recommend that you check:

1. Fuel pressure is correct and the low pressure warning light is working normally.
2. Check that the ignition system is timed correctly and all components are clean and serviceable.
3. Check the tune of the Spica system, with emphasis on specifically checking the operation of the Cold Start Solenoid (CSS).

[redblk63]

John,

Thanks for your reply. You are absolutely correct to suggest I try the simple/standard things first.

I am not sure about the nomenclature, we may be talking about different parts, or I may have forgotten, I thought the fuel delivery valve was part of the injector itself.

My query is about a "non-return" valve mounted where the injection line leaves the pump. (Sorry I do not have any of my manuals in front of me so I could use the correct terms.)

At any rate, YES, the pump has been resting along with a 76 Alfetta for years, and the gas had been contaminated by water. So it would not be a great surprise if these "non-return valves" (?) were leaking a bit.

The troubleshooting I used was to see that a little raw gas in the manifold would allow it to start and run long enough to pump gas up to the injectors, after which it runs and idles fine. (Could be CSS, I suppose, but the total lack of fire makes me think of these valves, would have to be all four of them.)

To answer your numbered questions:
1. Fuel pressure OK?
I really do not know, the pressure sensor does not work, and I have not yet tried my spare sensor, but it runs fine once started. (New main fuel filter.) Also, it squirts what looks like a normal distance from a disconnected supply hose.

2. Timing, etc.
Yes, timing is spot-on, and new spark plugs with hot spark. (I test spark with an old plug with the gap opened to 1/4 " to simulate the resistance of a cold, compressed, and rich mixture.)

3. SPICA tune, especially cold start solenoid?
Yes, I have done "by the book". This is my third Alfetta, and I have been pretty successful in tuning SPICAs up to now. In fact, I really enjoy the vast flexibility in tuning these devices. I do have high cold-idle RPM at the moment, which I will tune out once I get the start problem solved.

Excellent suggestion about the CSS. I recall I tested it for correct richening when energized, but I should recheck that.

With the help of experts, I'll get to the bottom of this soon, and report back with the solution.

All suggestions are welcome.

Larry in Pasadena, CA

[Roadtrip]

I'd suggest that you remove the pump from the engine altogether and drain the fluids from it and inspect for gas contaminating the logic section. The check valves in the fuel towers are usually pretty easy to remove unless they're corroded in there. If you don't want to remove the pump, you can remove the fuel towers and see what you've got in there. Also, no telling what kind of gunk and varnish is in the fuel pipes and injectors. That's why I recommend that if a system has been left to sit derelict, that no start be attempted until the entire system is cleaned thoroughly. I hate to say it, but it's likely the pump requires overhaul due to neglect and corrosion.

First things first though. The fuel supply system needs to be squared away before you do anything with the injection pump itself.

The high hot/cold idle could be due to a leaked-out T/A.

[redblk63]

John,

The hot idle is fine, cold idle is too fast, so I imagine the TA is doing its job. Lengthening the long-link rod might fix that, just a theory.

Yes, i did flush out the fuel system. Yes, I added Stabil after the fact. I even threw in a little two-stroke oil left from my last Bultaco ride to ensure good lubrication of the SPICA plungers for the first tankful. (I wonder why more people don't do that?)

I agree with the wisdom of your "remove it all" advice, but I am trying to be minimally invasive here for two reasons. 1. I am lazy. 2. I am not an MD, but try to follow the maxim "First, do no harm."

I'll pull off the pipes and have at the check valves from the top. Thanks for the council, and i'll report back after my next adventures. It may be a while, as other tasks beckon.

Thanks,
Larry in Pasadena

[George Willet]

Make sure that the cold start solonoid has power to it with the starter engaged. Sounds to me like it is not enriching the mixture during cranking. You can run a jumper wire from the battery positive to touch the contact on top of the solonoid while someone cranks the engine to test it's action.

If it fires right up then it may not be getting power from the ignition switch. I run a wire from the starter motor side of the starter solonoid to the cold start solonoid to engage it during cranking, thus eliminating the ignition switch and the added relays that were popular some years back.

[redblk63]

George,

Both you and John have recommended I test the cold start solenoid. I'd be a fool not to try that before beginning to dissect the pump, and will do so.

I'll verify it clicks with 12 V applied non-running, then monitor the voltage on its terminal during a start, and finally test its effect on the idle speed once the engine is running to verify it does its job.

I hope that is it, as I have a spare CSS.

Thanks,
Larry in Pasadena

[Alfacliff]

spam reported

[redblk63]

First off, thanks to the experts John Stewart and George Willet for their correct suggestions that the problem was in the cold start solenoid. I had tested that it affected the idle as it should when energized by a jumper cable, but had not tested that it was, in fact, energized at startup.

Upon testing that, I quickly added a jumper from the solenoid keyed terminal to the CSS. I am amazed at what a difference that makes.

Unexplained is why I could start before IF I cranked for a long time. Of course, I imagined the gas slowly filling the injector piping.

Oh well, all's well that ends well. Also, I figured out that if I removed the fuel towers from the injector pump, I could fish out the check valve using a length of nylon hose jammed over the exposed threads. My backup was a 5/16 pipe flange which could lossly thread onto the valve body, and then be pulled up with a screw and puller improvised from a washer and pipe. By the way, these check valves were somewhat mungy, and looked good after I lapped them with #280 (fine) valve grinding compound.

Starts right up.

Thanks again,
Larry in La Canada, CA