AFB accelerator pump not pumping with engine running...Argh!

[Aaron65]

I copied this from over at V8buick.com and the Buick forum here to see if there are any AFB guys who can give me a hand here...I had this thing apart four times yesterday...it's relating to my '65 Skylark...

All right, this is getting dumb. Basically, when the engine's running, I have no pump shot. If I put it on the bench, it's fine. Here's what I've done...

1. Replaced the inlet check valve. It didn't seem to work well without it.

2. Raised the driver's side float about 1/32", even though it was at spec, just to make sure the well filled. It should fill through the check valve though.

3. I found a couple of float seats that had larger orifices in them, in case the car was dropping the float level too much running (grasping for straws here).

4. I adjusted the pump so it started down in the well as far as I could.

5. The fuel pump has 6-7 lbs. of pressure to the carb.

6. The gas is way down in the pump well. I checked the pump discharge needle and it seemed OK, but I'm wondering if it's maybe sucking air. EDIT: I replaced the needle and there's no change.

The pump just won't shoot with the engine running (or at least it's very weak and sporadic). I'm just about out of ideas here. Maybe the carb has some kind of internal vacuum leak where it's sucking air in the pump system...Any ideas?

OR...anybody have a good 3826S core they could sell me? This thing's never run quite right with the current carb. It came with an Edelbrock, but the switch pitch doesn't match up to that thing very well.

[carbking]

A few things come to mind:

(1) Using the incorrect pump discharge needle (insufficient mass) of a pump discharge ball (still insufficient mass) could cause the charge in the passage from the pump to the pump jet to syphon (this might also cause the level of fuel to be low in the pump well).

(2) The original Carter pump 64-217s has a check ball (vent) internally in the pump. The purpose of this check ball is to allow air bubbles caused by heat which are created in the pump well to escape the pump well through the vent, thus preventing pressure to force the shot mentioned in (1) above pass the pump discharge needle. The ethanol in modern fuel will add to this issue, as will using an incorrect flange mounting gasket.

(3) The AFB accelerator pump circuit does not work well if a neopreme pump is substituted for the original leather.

Jon.

[Aaron65]

Thanks for replying Jon!

1. I'm using a brass-looking pump discharge needle that's of reasonable size, but I'm not sure what the original size one looks like. The current couple I've tried either came with the carburetor or in the Car Quest rebuild kit I used last. No discharge ball, just the needle.

2. The lack of discharge occurs even with a cold engine. I'm using a replacement pump that likely doesn't have the check ball. I'm using a sandwiched aluminum/gasket/aluminum/gasket flange gasket setup by Mr. Gasket that insulates the carb somewhat from any heat from the engine. The car doesn't have a heat riser anymore, but the passages aren't blocked off either, so the gasket does get sooty. The car never has a vacuum leak, however.

3. The pump I'm using is leather (I've tried two).

[NTX5467]

I would HIGHLY RECOMMEND that you get a solid carb base gasket -- period! Using the beloved Mr. Gasket insulator "stack" gasket set, I discovered that if you don't torque the carb retaining nuts down very evenly AND use a good bit of discretion in how much torque you use to tighten things down, you'll end up with a cracked baseplate area on your carb. In my case, it was adjacent to the ported vacuum port on the rh front corner of the carb. NEVER AGAIN! (plus a few other "nice" words when I found the crack).

The stock float seats should be fine for your engine. Unless there's been some extensive modifications or the carb under the engine has morphed into a much larger displacement, the larger diameter orifice seats are not needed, but they won't hurt anything either. Just as raising the float level on the pump side would have no affect, either. But I understand your reasoning in doing these things, too.

The anti-siphon check valve is probably just under the accel pump shooter. On a ThermoQuad I had on one of my Chryslers, I was also using a sodium-based valve seat recession additive in the fuel. It seems that after the car sat, the sodium-based additive caused that check weight to stick in its bore, in the "no flow" position. Everything I did would not get a pump shot, no matter what I did. It made the car very hard to start, too. I checked the linkage and it was moving the plunger. There were no leaks in places, either. When I got around to further checking things out, I took the shooter off and discovered the stuck closed weight. Once I freed it up, things worked normally again. Never used that additive again, either.

I would take the pump shooter off of the venturi area. Carefully, so as to not drop that hold-down screw! Then see if the anti-pullover valve is free in its bore. If it is, then carefully remove it and check for possible burrs on its edges, but I also highly suspect you'll not find anything there, unless it might be some residual fuel-related coatings. Then, with an accel pump plunger, gently depress it in the bore (carb top removed) and see if any fuel bubbles out of the pump discharge circuit into the vehturi area of the carb. If it does, then put the weight back into its bore, reassemble the pump shooter to the carb, and retry the pump shot.

One of the many Carter kits I have used to mention, in the process of rebuilding one of those carbs, using the existing (i.e. old) pump check ball, using a small punch, tapping the punch (placed on top of the existing pump check ball) and giving the punch a few taps to ensure there was an acceptably round surface for the new ball check to seat against. I never did do that, it would seem that only a really corroded or degraded surface might need this extra little bit of work.

If, after removing and then replacing the anti-siphon weight, no pump shot, then you might also need to check the pump shooter nozzles to see if they're open. You can also visually do this when you remove the shooter initially.

The pump circuits on those Carter carbs are pretty simplistic in design. I don't know of any nearby places the pump circuit might be near a crack-prone area, either.

One last question . . . when you activate the pump plunger, are you moving it slowly or reasonably quickly? It seems that, from my experiences, there might be enough wear on the accel pump bore that it might compromise the ultimate output from the pump, unless you move the plunger reasonably quickly. I also believe the pump bore tapers, such that if you adjust the pump for too long of a stroke, it might start in the looser, upper area of the bore, which might make getting a solid pump shot more difficult. If you've tried to adjust the pump lever link, for any reason, I suggest you return everything to stock settings.

The other thing is that with a Holley accel pump (4150 style, plus relatives thereof), any throttle movement can result in a pump shot. On a similar Carter AFB (or relatives thereof), a small movement might not result in ANY pump shot, for various reasons . . . but they always seem to respond just as well. Perhaps it's the metering rods taking up the slack in that department?

Please keep us posted . . .

NTX5467

[Aaron65]

Thanks for the great ideas! I have a short spacer I can try in place of the stack, but I always just snug down carbs by using my hand on the top of a ratchet. That doesn't mean they won't crack! The thing that's turned a light on in my head is using the punch to seat the pump check ball (or in my case a needle). That was one of the things I was wondering...was the circuit sucking air in from that area. More to come, eventually. Thanks again!

[NTX5467]

Of all of the carbs I've done over the years, myself or others have NOT done that "re-seat" procedure . . . nor have the new items failed to operate as designed. I classify that as one of those little things that might make things perfect, but then they are already plenty fine as they are. And, of course, you know what happens when some people get hammers and punches in their hands!

Personally, the "sucking air" or "leak" issues are very improbable, due to the way the carb's circuits are designed and situated in the casting. That's my observation on those issues.

Just some thoughts,

NTX5467

[Aaron65]

I think we're at the improbable problem stage. It works great on the bench, so I know there are no blockages. However, as soon as the engine's running and I work the throttle, nothing. If I disconnect the throttle rod and work the pump manually, I can get a shot but I can tell the well doesn't refill correctly because I can feel the fuel level in the well go down every time I work the throttle. I won't take a giant whack with a hammer and punch...just a couple light taps.

[NTX5467]

Has this car/carb been sitting for a while? Reason I inquire about that is that if the carb has been sitting long enough for the fuel to dry out of the carb, with a bit of ethanol in the blend (even if it's not the full 10%), plus some moisture over time, there might be some "corrosion" in the passageway, not enough to clog it, but enough to restrict it a little? I might need to do some research on the circuits and if there might be any "ball-sealed & soldered" drilled passageways which might be openned up, clean out, and then resealed/soldered.

What about fuel pressure? Same issue with the vehicle sitting long enough for the fuel to evaporate/drain from the fuel pump . . . AFTER it's had ethanol'd fuel in it. The ethanol (remember, "the great cleaner" ethanol is?) will dry out the rubber's "oils", which will make it hard and somewhat porus at the same time. That's the issue with both rubber hose AND fuel pump diaphrams.

Just some additional diagnostic thoughts,

NTX5467

[Roger Zimmermann]

I had this problem once. Take the plunger out, take a small screwdriver and expand gently the leather. You will see, it makes wonders!

[Aaron65]

Roger, you are an amazing genius and I owe you all the thanks in the world! That's all it was...I now have pump shot every time. Thanks for sharing your experience! And thanks to NTX and Jon for bearing with me...

[Ben Bruce aka First Born]

SO, the leather was just shrunken? Who would a thunk.? Glad it is working now.

Ben

[Aaron65]

Not really...I tried two leather pumps from carb rebuild kits. They're just not quite expanded enough out of the box to make a good seal in the pump bore, or so it seems. Kind of a weird deal, but I'm glad it works! You learn something new every day!

[Roger Zimmermann]

Roger, you are an amazing genius and I owe you all the thanks in the world! That's all it was...I now have pump shot every time. Thanks for sharing your experience! And thanks to NTX and Jon for bearing with me...

Me, a genious? Not at all! As I wrote, I had hesitations during acceleration with my '56 de Ville. A friend of mine gave me the tip; I'm just telling it further. I'm glad it helped!

When manufactured, the leather is shaped probably into a press; the leather wrinkles are sticking on each other. The cardboard cylinder put on the piston is there to avoid damages and that the leather is expanding by itself in the box. When the pump is ready to be used, the fuel and the spring pressure are insufficient to free the wrinkles; a little help is needed...

Edited November 22, 2012 by Roger Zimmermann (see edit history)

[NTX5467]

Thanks for the update. I remember guys overhauling carbs back then taking the accel pumps with the leather/spring plunger and rolling the leather backwards upon itself (upward) to "rough it up" before putting it in the carb. That was a common thing to do, back then. I suspect the ethanol content might now tend to dry the leather somewhat?

When we had one of our first hostings of a BCA Board of Directors meeting, in the 1990s, in the hospitality room, one of our members had just got his '55 Roadmater 2dr ht going. The carb had an accel pump issue which nobody seemed to be able to fix. They'd rebuilt the carb several times, to no avail. Even sent it off to a rebuilder, with no improvement, either. One of the BOD members from AZ suggested he get the original leather pump, flex and rough it up, and put it back in the carb. He'd had a similar issue with a '50s Pontiac with the similar carb on it. Fixed that one, too!

Thanks,

NTX5467

[Aaron65]

Kind of makes you wonder how many old cars are running around with a similar problem...

[carbking]

Probably lost knowledge:

When installing a leather accelerator pump, first lubricate the skirt by saturating in a light machine oil (3 n 1, sewing machine oil, etc.), then gently expand the skirt as suggested by Roger. A very small Phillips screwdriver bit works well.

The same technique may be used to rejuvinate a dried out leather pump. 99 percent of the time, a 20 year ole (or even older) leather pump rejunivated in this fashon will function better (an longer) than one of the cheap modern neopreme pumps (regardless of the color of the neopreme).

Of course, if the leather skirt is torn, then the pump is junk.

Jon.

[Guest cben09]

The pump leather will like Neatsfoot oil better,, Use the same on clutch leather !! OOps ,,,generation gap,,haha,,,Ben