Stalls on decelleration and when nose is down

snip...

Could be as simple as a vacuum leak?

Not the same engine as yours but my 2.0L type 4 engine did exactly what you are describing, turned out to be two little rubber elbows that had split on the pipe between the inlet manifolds.

I'll be darned. I guess this diagnostic limb I went out on was strong enough to hold me without breaking: the reason the Wonderbus's engine is dying when the car is braking or sitting with the nose lower than the rear does appear to be low fuel pressure, like I guessed.

For single-carb engines, Bentley's says we want 3 to 5 psi. The pump is putting out 2.8 psi. Braking and nose-down conditions pull fuel toward the front of the bus, away from the pump and I guess the pump does not have sufficient sucking power to keep it, er, up.

Next step: identify the cause of the low pressure. The bus accelerates and cruises normally, so I don't think I have a restriction anywhere. Bentley's says that the pump pushrod is supposed to extend 13mm (.511'') beyond the upper face of the Intermediate Flange (Bentley's Type 2, Fig

4-3) at its maximum extension for proper pump operation. Mine's at 12.1mm (.476'') maximum. That's with NO gaskets between the Intermediate Flange and the engine.

I recall reading someplace that one way to lengthen the rod is to weld a little bead on the end and polish it . . . but I don't weld.

I do, however, have a belt sander. How about taking a little off the top of the Intermediate Flange? (Bentley's Type 2, Fig 4-3).

More better would probably be to buy a new pushrod . . . but what kind of fuel pump does this engine have. Single-carb 1600 DP, so it is not a side-mount. It does NOT exactly match up with any of the illustrations in Bentleys. It mounts on top of the engine like Fig 4-3, but it does not look like Fig 4-3 . . . the top has four bolts like the 4-2 fuel pump. And it's not the Fig 4-4 pump.

While the pump is out, I'll disassemble it and clean it and look for problems there.

Am I sorta/kinda on the right track to solving this problem? Is the belt sander Intermediate Flange modification, like, really stupid?

-- Mike Rocket J. Squirrel Elliott

Sounds good to me. I don't know about your solution, but if the bolts tighten up afterwards, why not? You may have to finish it wet sanding on a piece of flat glass (with wet/dry paper of course) I do that in the sink with water trickling continuously. (It must be perfectly flat and the two surfaces must be somewhere close to parallel.)

Thanks for the feedback. I think that my diagnosis must be incorrect, though. Low fuel pump pressure should not cause the engine to die when it is returned to idle. Someone who knows a lot more about these things than I ever will e-mailed me to point out that there is enough fuel in the carb bowl to let the engine run for a minute or so.

That notwithstanding, I am wanting to get the fuel pressure up to the specified 3 to 5 psi range. I hoped that a longer pushrod would do the job -- the 110mm rods from Jbugs.com (within driving distance) looked perfect. But they actually measure at 106mm, and mine is 105mm. Their shorter "100mm" pushrod measures 100mm. Maybe some other vendor, like aircooled.net will have a longer pushrod.

Anyhoo, the Smarter Guy suggested that I probably have a vacuum leak in the brake booster. This does line up with my experience that the engine dies when returned to idle while I am either decelerating, or when the bus is sitting nose down at a red light. In both cases, I am using the brake.

So I'll paste this thing together and check to see if the engine dying is related to brake usage.

Okay, fuel pump is back on (slightly longer pushrod, but still disappointingly low 2.8 psi). While idling, there is no effect on rpm's when brake is applied. On a short drive (to the grocery store -- Mrs Squirrel wanted me to pick up some mac 'n' cheese for our visiting grandson) I could not get the brakes to cause the idle rpm's to shift upward or downward at all.

Except twice. Two time the engine just died when I was braking. Before those two times, and afterwards, zero brake-engine interaction. But nothing else seems to cause the engine to die. That said, I can't even be certain that the brakes are causing the engine to die, or if it was just waiting for my foot to go off the gas to die.

Either way, the engine started right up again.

Sheesh.

You're running a Pertronix ignitor, right? Try a vacuum pump on the diaphragm while running. Pump it up, release it, etcetera. See if there is any missing or cut out related to rotation of the breaker plate (vacuum advance.) If so, look for a wire shorting or a wire being bent sharply as it exits the little black box (I'm picking right up on that electronic jargon aren't I!) Or a grounding problem. I had one with which the wire was being bent sharply as it exited the housing which caused the strands inside the insulation to break even though the soft insulation was still perfect. Anther common one is the grounding wire that keeps the breaker plate grounded, breaking.

Another diagnostic tool: If you have a timing light that doesn't require house current, hook it up to the coil and position the light such that you can see it in the rear view mirror. When the motor conks, glance in the mirror and see if it's still flashing. That way you'll see if it is fuel related or ignition related. (Your bus has a top hatch in the engine compartment doesn't it?)

You're right about the fuel-in-the-floatbowl situation.

I wonder how your check valve is on the vacuum line to the brake servo. Maybe it's sticking intermitently?

Intermediate

Compu-Fire pointless and Tiger CDI. But I am not getting this vacuum pump on the diaphragm thing. Makes me think of Austin Powers and the penis enlargement pump . . . but I don't think that's what you are talking about at all.

Wait -- I see, you're suggesting I look for a problem with the points replacement module's wiring. Good idea, though that vacuum pump thing I have to think about. Maybe I can modify my bike pump? Nah.

Nope. I can figure out how to run the wires from the engine hatch into the rear of the bus anyway. But, uh, when the engine conks, wouldn't I pretty much expect there to be no spark? How will the light give me any information?

Yeah, well, if I'd ever rebuilt a carb I probably would have known about that.

Hm. I reckon the only test for that is to R&R the check valve.

Thanks! And it's nice of you to help!

-- Mike Rocket J. Squirrel Elliott

The vacuum diaphragm operates a lever attached to the plate the breaker points are mounted on (until you remove them and insert CompuFire!) The movement of that plate (rotational) is what retards the timing on a pull (low vacuum) and advances it when engine's not working so hard (high vacuum). It is not uncommon for there to be intermitent open circuit problems for various reasons (as I mentioned earlier) while that plate rotates.

The light will keep flashing until the engine stops turning which is likely several seconds after it cuts out. Perhaps "conks" was improper use of the phraseology. If the engine cuts out and you look in the mirror and the light has stopped flashing, you can bet on one of two things: The problem is not fuel (OR) You didn't look quickly enough and the engine had stopped rotating! (Unlikely)

Didn't you put a new check valve >

accelerates

illustrations

One more vewwy important item: Try running the CDI on bypass mode - If yours is as mine was, you simply turn the plug over and it bypasses the CDI and runs your conventional ignition. Might as well try to eliminate the item that is most likely to fail right at the start! -BaH

May or may not help, but when I had this problem with the vacuum leak, pressing the break peddle at idle made no difference to me either, and it also used to just die when coasting down without really touching the breaks, with mine I guess as the pipe I had a leak in was after the carb, it was something to do with too much air in the mix when slowing.

Makes sense! Now, back to that "vacuum pump" thing . . .

I dunno, when the engine dies after I take my foot off the gas the needle on the tach drops like a stone. "Conks" is a pretty good word, I reckon. The trick I suppose is to keep an eye on the timing light whenever I take my foot off the gas to see if it just goes dark or not.

Some other rodent or marsupial. I'm getting a new check valve now.

-- mrjs

The answer is Yes, you CAN adapt the pump from the aparatus that Mrs. Squirrel gave you on that second date waaayyy back when... (and she married you anyway!)

"> Makes sense! Now, back to that "vacuum pump" thing . . .

I can just put a reducer into the chuck. I had to enlarge the orifice from the stock diameter to nearly 2-1/2 inches when I got the pump.

Okay, it's apparently not an ignition problem. I've got my timing light securely taped to the bike rack on the back of the Wonderbus. Gives a delightful haunted house, or 60's psychedelic light show effect at night, I might add. The engine died a couple of times just like before (take foot off gas after revving a bit and engine conks) but only twice out of many such moments. The timing light did not go dark suddenly, but flashed slower and slower as the engine spun down. So it ain't ignition. Started easily.

The reminder that I can bypass the CDI by reversing the plug was a good one, but I don't think I need to look any farther in that direction.

It does not appear to be a fuel problem. I was reading 2.8 psi, lower than the Bentley's spec, and a longer pushrod did not bring it up any higher. That said, a got a second gauge yesterday to see how much measurement error I might be getting -- a fine, highly-uncalibrated $19 Sears piece of instrumentation -- and it read 3 psi. What my actual pressure is is uncertain . . . neither of these gauges are calibrated. But the engine runs fine at all speeds and even if the fuel pressure dropped suddenly, the carb has enough fuel in the bowl for idling for a minute or two, as has been pointed out.

This intermittent conking is not influenced by engine temp -- hot or cold, it can happen.

Intermittent brake booster check valve? I have one on order from aircooled.net and will see if it makes a difference.

I appreciate all the clever diagnostic methods that y'alls have been offering.

== Mike RJS "Of the famous pencil > >

First name "Wood"

A bit later. It's not brakes, near as I can tell. This morning the engine conked a few times so I was able to get a better feel for what's happening.

1. When it happens, which is about 5% of the time, it happens at return to idle.
2. It does not require braking to conk.
3. Timing light continues to flash a few time in synch with spindown, so I don't think its ignition.
4. It never happens at return to idle if I have just been pooting along the street at low rpms.
5. It generally happens at return to idle after the engine has been pushing at higher rpms for a few minutes. Freeway offramp, for example. Clutch goes in, brakes not used, conk. OR, today, it occured after climbing a hill, about 1/2 mile, right turn onto level street without slowing much, on the flat, clutch goes in to downshift and conk. Or once just between shifts while crusing at 45mpg on city street. I went back to the bottom of that 1/2 mile hill and re-did the route to see if it would conk again. It did not.
6. Engine can be hot or cold.
7. Restarts easily and idles fine.
8. Fuel pressure measures 2.8 to 3.0 psi, depending on which gauge I use.
9. Other than this conking problem, the Wonderbus drives very, very normally.

John Connoly e-mailed and suggested that I check to see if the carb is adjusted too rich. Will check.

== Mike "Pencil in the cylinder" Rocket J Squirrel

Hey Mister Squirrel,

Have you considered that the clutch or brake pedal are shorting something out when they are depressed? Perhaps there is a loose wire that wiggles about after high engine rpm or when facing downhill?

I once had a problem that the engine would conk out everytime I put it in reverse. Turned out to be a short in the backup light circuit.

Fweem.

Dave T.

Well . . . the timing light indicates that the engine is still getting spark on its way to conk-dom.