Mikuni 44 Accel Pump Stroke... Wrong spec in manual?

[inline6]

I have two brand new 44's and two older ones that were on my 510 years ago. Looking at the Mikuni 44PHH manual and the Honsowetz book 'How to Modify your Datsun/Nissan Engine' (page 113), both call for 7.5 mm of stroke at the accelerator pump lever. Measuring the new Mikunis, which I assume were assembled correctly, as well as the old ones, I am coming up with only 3-3.5 mm of stroke for each of the four carbs. All of them have the cotter pin in the middle hole. All four have 2 thick washers stacked on the rod above the lever and the spring. Seems like taking those thick washers out and putting just one thin one on either side of the pump lever (like the manual shows) would be in order. Then adjust the pump rod length as necessary. Then again, maybe the 7.5mm spec is incorrect. Any thoughts on whether having these things set up for 3-3.5 mm is right?

[jmortensen]

In general terms I think you want to do less with the pumps and more with the pilots. I tried changing the lever and running huge pump nozzles to fix a stumble at about 2500 rpm and it just got worse. What made it better was running larger pilots, and after talking to Rebello and Malvern and Wolf Creek Racing I was left with the notion that the pump nozzles should be 40's and the stumble should be overcome with pilots or smaller chokes.

I never measured or checked the throw on my pumps so I can't speak to the measurement question directly, although I did change the diaphragms thinking that was the cause of the stumble, to no avail.

[inline6]

I read the post(s) about stumbling with the long runner Cannon manifold. I am in the process of switching over from SU's to Mikuni 44's on the Cannon and I know I am not going to be happy with a stumble :mad:

I am curious... Why would the Mikuni service manual specify such an incorrect specification? I hear what you say about the pilot jet and what you experienced. In a futile attempt to think through this logically, I can't understand why that would help.

There are two things going on with the pump. The pump nozzle regulates the duration of fuel injection as does changing the length of stroke on the pump rod. According to the manual, the larger the nozzle bore, the shorter the duration of the fuel injection. And the longer the stroke, the longer the duration of fuel injection. Seems logical that by playing with the two, you could come up with longer/shorter duration fat or skinny streams of fuel injection (you evidently had the same logic or you wouldn't have tried what you did). And if these carbs are set up with 3mm of stroke instead of 7.5, then won't that greatly shorten the fuel stream injected - a potential cause of the stumble? Maybe what is needed is a really long duration, skinny stream of fuel... That is longer stroke, smaller pump?

Also, according to the service manual, the pilot (fuel) jet is functional at low and medium speeds, or in another place in the manual, only up to 20% throttle opening. People have said that they get a stumble when at say 2-3k RPM and they punch it. Again, according to the manual, a lean condition is created because fuel being heavier than air, when the throttle plates open, the flow of fuel will lag behind the flow of air. And, once you kick the throttle plates open all the way, the pilot system should be out of the picture...

Not sure why what makes sense on paper doesn't work in real life. I plan to buy a narrow or maybe even wide band O2 sensor and try to go about this as scientifically as possible. Did you use one when you were making changes?

By the way, Dave from Malvern was a friend. I grew up in Charlottesville. I bought a hot rod 510 in college that he helped me with. He had me install a 490/290 Nissan comp cam in the L-20B 2.3L which also had 44's. Changes he had me do to the carbs? 34 chokes were too small for that displacement he said. Install 40mm chokes. Also, he said I needed larger pilots and pump jets. Pilots ended up being 57.5 and pumps went from 35's to 50's.

[jmortensen]

I read the post(s) about stumbling with the long runner Cannon manifold. I am in the process of switching over from SU's to Mikuni 44's on the Cannon and I know I am not going to be happy with a stumble :mad:

I never tried that guy's suggestion of going to the shorter manifold, and it never made sense to me why that would help. My understanding of what is going on is that when the throttle is snapped open you have a loss of velocity in the runner. If you use a smaller choke, the velocity is higher, so that minimizes the problems. If you use a shorter manifold of similar runner diameter I don't think the velocity issue changes too much, but like I said I never tried it.

I am curious... Why would the Mikuni service manual specify such an incorrect specification? I hear what you say about the pilot jet and what you experienced. In a futile attempt to think through this logically, I can't understand why that would help.

There are two things going on with the pump. The pump nozzle regulates the duration of fuel injection as does changing the length of stroke on the pump rod. According to the manual, the larger the nozzle bore, the shorter the duration of the fuel injection. And the longer the stroke, the longer the duration of fuel injection. Seems logical that by playing with the two, you could come up with longer/shorter duration fat or skinny streams of fuel injection (you evidently had the same logic or you wouldn't have tried what you did). And if these carbs are set up with 3mm of stroke instead of 7.5, then won't that greatly shorten the fuel stream injected - a potential cause of the stumble? Maybe what is needed is a really long duration, skinny stream of fuel... That is longer stroke, smaller pump?

On the issue with the pumps, I have convinced myself of the following (not saying I'm right, there are many different opinions on this stuff):

1. The pilot affects a lot more than the manual says it does. It will be contributing to the overall mix pretty high up in the rpm range, 4-5000 rpm easy. I like the "How to Modify" suggestion of 1.5 turns out on the pilot screws. If you need more, get a bigger pilot, less get a smaller one.

2. The pumps squirt a big pee stream of gas into the runner. Take the carbs off the car and open the throttle and you'll see what I mean. If there is little velocity, that pee stream of gas isn't going to atomize as well and that is where I believe the stumble comes from.

3. A richer pilot will help to fill the gap that the pumps don't fill well.

4. If you could get different jet blocks, that would be the way to solve the problem. I was planning on trying some of the OB or 8 blocks to see if they helped but never got around to it. Webers have tons of different emulsion tubes to choose from, Mikuni owners don't.

Is it possible that a longer stroke could help? Sure, if I'm wrong about the velocity in the runners thing or if it could be thin enough to atomize properly and fix the lack of fuel in that one rpm range.

Here's a post from another forum about getting the mains to come in earlier. Looks like that's a function of the emulsion tubes, so at least according to this forum I may have been onto something with the jet block thing. It does seem like getting the mains to come on earlier would negate the need for a larger pilot. The fat pilot is definitely a crutch: http://www.innovatemotorsports.com/forums/showthread.php?t=2309

Also, according to the service manual, the pilot (fuel) jet is functional at low and medium speeds, or in another place in the manual, only up to 20% throttle opening. People have said that they get a stumble when at say 2-3k RPM and they punch it. Again, according to the manual, a lean condition is created because fuel being heavier than air, when the throttle plates open, the flow of fuel will lag behind the flow of air. And, once you kick the throttle plates open all the way, the pilot system should be out of the picture...

Play with the pilots and I think you'll find that I'm right that they do a lot more than what the manual says. I was tipped off to that fact by Dennis Hale and I found him to be correct on that one when I played around with it. YMMV.

Not sure why what makes sense on paper doesn't work in real life. I plan to buy a narrow or maybe even wide band O2 sensor and try to go about this as scientifically as possible. Did you use one when you were making changes?

I used a narrow band with a $6 Harbor Freight voltmeter and shot for .8V at WOT. I had it set up on the console and used it constantly. The refresh rate is a problem, but I couldn't find an analog gauge for 0-1 volt. I've had several friends buy the Autometer gauge and need to replace the crap O2 sensor that comes with it immediately. The gauge would just be full rich or full lean, no middle ground at all. I used a Bosch one wire for a 90's Chevy on my car, but friends have used the heated O2 as well with the gauge with good results. The gauge (with Bosch sensor) was given the seal of approval by a friend who had his car on a dyno with a wideband in the tailpipe. His story was that the dyno operator was laughing about what a POS the gauge was until he saw for himself that it was right on with what the wideband showed.

By the way, Dave from Malvern was a friend. I grew up in Charlottesville. I bought a hot rod 510 in college that he helped me with. He had me install a 490/290 Nissan comp cam in the L-20B 2.3L which also had 44's. Changes he had me do to the carbs? 34 chokes were too small for that displacement he said. Install 40mm chokes. Also, he said I needed larger pilots and pump jets. Pilots ended up being 57.5 and pumps went from 35's to 50's.

Interesting. It was a long time ago, probably coming on 10 years, so I could just be remembering his advice incorrectly. I'm sorry he's not around to ask. I know I was told to go from 34's to 37 choke tubes and I tried the larger pump nozzles and then was told to go back down on those. It ran great at higher rpms but had the stumble so the next thing I was going to try was smaller chokes to fix the stumble, but I was autoxing so the stumble was more of a problem than the hit to the top end would be. I had reduced the stumble A LOT from when I started. I'd say it was probably 85% fixed compared to day one.

Edited September 9, 2009 by jmortensen