A Tale of Two Carbs, Part 1

Carter YF Rebuilt, Lessons Learned

by Dave Pete

In 2015, Dave decided the best way to rebuild the Carter YF carburetor on his 1954 Jeep CJ-3B "Lil Willy" was to really understand how it worked, and then he undertook to make the lessons he learned available to everybody on the CJ-3B Bulletin Board. Here is Part 1 of his report, with some useful comments from readers. Part 2 will go into detail on the Lo-Speed Circuit and Hi-Speed Circuit. -- Derek Redmond

The Five Circuits

When we lived in Fairbanks, AK, I tried to fix up Lil Willy's carburetor, which was the better of my two units, but both had seemed overly rich - evidenced by gassy smells, poor performance and a wonky idle, and by sooty spark plugs! And my second carb and extra parts were back home in Wyoming.

As it turned out, I reassembled that carb and made do until today. Since we got back home to Wyoming, many things have floated down that river of life and under my bridge, but here recently, it just felt like the right time to do something, especially when I saw oldtime's excellent ability and carb knowledge coming through on the CJ-3B Bulletin Board.

I got the carb off the Jeep, and I gathered up all my various carb parts from the other unit, including multiple little ziplock baggies full of grouped parts, and started reading.

I first want to say as loud as I can, "Thanks so much to Derek, oldtime (Ken) and so many others who have both shared knowledge and have compiled such excellent information for all of us hobbyist Jeepers".

I read up on the carburetor under "Group 12: Fuel System" on Specs and Tech Tips. Then I read what Carter YF posts I could find in the 3B forum, including doing a forum search under "Carter YF". After days of reading, and disassembly of my carb parts, and regrouping like parts and verifying measurements, calibrations and such (I bought a micro drill bit set and broke out the calipers), I had acquired and re-acquired enough retrievable knowledge to actually understand the process this time.

I've built many various carbs over the years, but I've been good enough at it mainly because I can follow instructions fairly readily. But really understanding the "circuits' has seemed too difficult to me, so I haven't applied myself to that area, until now.

Oldtime said something (to somebody else in another post) along the way that resonated with me this time, it was something like, "in the meantime, study up on the five circuits".

So I did.

This post is not going to be "how to built a Carter YF". I'm not good enough to attempt that, especially when we are so fortunate to have others like oldtime (and I have to specifically mention jyotin and scoutpilot) with great knowledge and experience on the subject. This post will cover specific issues about what I found on the Carter YF, including pictures that I hope are beneficial.

I hope those who know better than me will speak up and correct mis-statements I may make. I can usually take in a great deal of new (or old) info and use it to advantage, but then I usually forget stuff fast when I'm not working with it. Already, two days after completion, my mind is getting foggy on certain aspects of what I just did. I think we're all different in that way, how much we take in, how we use it and then how we lose it. So please, don't anybody allow something I say wrong to stand for long, let's get it corrected when you see it!

I learned these basic, and particular, issues for our Carter YF carbs.

Fuel Supply Circuit

This deals with the float bowl fuel. The inlet orifice, closed by the float fuel shut-off needle, must be smaller (orig .076, but no larger than .090 and apparently .088 is available in the Hygrade 518C kits) than what comes in most rebuild kits, because the carb was used on many types and sizes of engines. I highly recommend a $4 micro-bit drill bit set from Harbor Freight or such someplace (in fractions of millimeters, but also package labeled in thousands of an inch) so you can make good estimates of orifice sizes.

Our little four cylinder engines cannot take such a great amount of float bowl fuel like those big inline sixes, so we have to be very careful to retain our serviceable parts, as opposed to the natural tendency of discarding old parts, installing shiny new parts from a kit and hating life ever afterwards. Oh - and if we take our carbs into a shop so THEY can do the hard work, we need to somehow prevent THEM from discarding our correct and serviceable parts and installing "ever after problematic parts because they were new and shiny". I think it might be correctly stated that this is the one main reason why so many people have a problem with the Carter YF in their Jeeps! This is the main condition that causes a "flooded Carb" after a rebuild. And that is what seems to drive MOST of the CJ3B posts about Carter YF complaints!

One last thing on this circuit, the needle. We're supposed to have a needle with a spring and a pin (three parts) as opposed to a solid needle hanging from a clip attached to the float. The point of this is because of our rough duty (off-roading).

Detailed pictures coming on this issue. But first, next circuit.

Choke Circuit

There's nothing really to discuss here people. Manual choke like mine, or later years with automatic choke. Oh sure, there's some nuances here, but I didn't find any needing mention. People don't tend to threaten to sell their Jeep or install a Solex carb because of choke issues. Let's move along.

Lo-Speed Circuit

Now this was a very interesting circuit to think about. In general, this is your idle and up through about 20 miles per hour. There are a whole lot more orifices (for fuel flow and for air bleeds) in this circuit than I realized. And obviously the fuel and air bleed flowing though this circuit varies in great amount between idle and 20 MPH! But - the main thing to think about here is making sure the carb parts are cleaned correctly (soaked in the Chem-Dip carb cleaner 24 hours) and the passages blown out with ample air pressure. After soaking, I wash in VERY hot water, drain on a towel and then dry and blow out with air.

You don't really get rebuild kit parts that mess you up in this circuit. Having said that I would recommend a very careful attention to gasket use and placement so as to not cause yourself a problem here. Take the time to Google pictures of the circuits, so when you are attaching carb bodies to one another (e.g. fuel bowl body to throttle body), you aren't inadvertently covering up holes and blocking passageways that you just so meticulously cleaned out with compressed air! In other words, know what those tiny holes are for. Follow them.

Hi-Speed Circuit

Messing around here is rocket science (as oldtime implied). If you just love mathematics, you can have a ball here. Since so many of us live and play at altitude with our Jeeps, and because we can't just run down to the corner auto parts store and pick up a "high altitude kit" (metering road and main jet) like we once could say... oh fifty years ago, we are pretty much stuck with the rod/jet in our carb. Almost all of these are some variation of standard (surface to 5000'). But - the extreme fine tuning MAY not be as important as we thought. We have other issues to make the unit work, like the above flooding thing for instance. Fine tuning the jet and rod for that San Juan Mountains trip to Colorado may not be all that important after all. But in the case that you get everything working fine on standard, and THEN you want to fine tune, I'm going to show some pictures and discuss a little "fancy stuff" toward the end of this thread. You see, I have what I think MIGHT be a high altitude kit in my possession that is not only made aftermarket, but might be two steps lean for one or the other of the Jeep Carters (I suspect the 938SD).

But for now, let's continue with more important issues, like this next circuit.

Accelerator Pump Circuit

The ubiquitous "stumble" or flat spot upon acceleration. Carter tried a number of things over the years to correct this issue. It appears there was originally a check disc used in the circuit, along with a weight (Bulletin - 1951). Note the bulletin shows replacement of a disc with a ball, but that it is for carb numbers 738S, 768S & SA, and 832S. More applicable to CJ2A and 3A perhaps? It's found in Carter YF Carburetor Documents on CJ3B.info.

By 1960 the Service Manual (pages 102-103) was showing an upgrade for the 938S, SA and SC from a check ball to a check needle, as found in new 938SD's! It is THIS service manual procedure which also discusses the need to drill out the "pump jet" from .025 to .031 only WHEN using the upgrade kit to a metering rod and jet found in a 938SD (75-1500 and 120-166), and that's when you re-stamp the carb as a 938SD.

Interestingly, that procedure says to discard the retainer clip (which oldtime says elsewhere isn't a problem to use and it just might save you from losing your check valve). oldtime talks about the sixty second test, and these things are covered in great detail under the main page Tech Tips and the multiple threads in the forum. I won't expand much on this, but I will say I think many conversions may have failed for a variety of reasons. I'll talk about these and show detailed pictures when we get to this item.

However, I discovered something in this circuit that might well cause a MAJOR problem for folks doing a modern rebuild on these old carbs. That is to say a poorly fitting gasket between the fuel bowl body and the throttle valve body, might very well shut down your vacuum assisted portion of the accelerator pump design. Would that cause a stumble? Damn Skippy! to quote a very colorful fellow also named Dave whom we met while living in Virginia. Apparently that's an eastern way of saying "yes"! Pictures to come.

Rebuild Results

I ordered a Hygrade 518C kit, as oldtime recommends if you cannot locate a Carter OEM kit (which is getting harder almost by the hour I think). But the kit didn't come yet. I reused gaskets and existing parts. It worked out VERY well. But I don't usually throw away the old parts so I had some extra stuff on hand, including two or three extra gasket sets in various states of repair. I'll keep the new kit, when it comes, for a future use, because you know we will need to. Maybe!

For now, here are the spark plugs after a test drive up to the mountain overlook of our city. Prior to the test, the plugs were quite sooty, dry, fluffy and black. Before testing the new carb, I simply brushed some of the soot off with a tooth brush (it only tastes bad for the first brushing afterwards). And you can see from the picture, the plugs are already turning a nice healthy tan!

user posted image

Why? I'm using standard (sea-level to 5000') jet and rod for a 938SD (rod 75-1500 jet 120-166). Probably because my city elevation is about 5000'. We wheel a little below that and quite a bit higher, so that will vary, but at least I don't have the "rich" condition that until now is the only condition Lil Willy has known under my ownership.

I mentioned my "two carbs". The one that came with Lil Willy was a donor engine (he had lost his block and crank and much of the rest of his original engine components including the carburetor). That donor engine, I was told, came from a Forward Control Jeep, and the engine was painted blue. There is still blue paint visible on the cast iron throttle valve body of the carb. It is a 938SD, and it is in pretty bad shape, but that's where I got the rod and jet I'm using now.

The other carb came on another donor Jeep I bought a few years later. That was a 1953 (one year older than Lil' Willy) but the carb was a 2756S. In the initial link of this post, oldtime identified that carb as a Kaiser Industrial application. I came to find it was in essence, a 938SA (in Jeep speak). I concluded that because it used a standard rod and jet for a 938SA (rod 75-996 and jet 120-155) and the air bleed orifice for the accelerator pump check ball was .025, while the 938SD is a .031. (If you read my Fairbanks post you'll see how I originally said that was .031, but I was wrong at that time).

So in Fairbanks, I had the right jet/rod in the wrong, albeit in better condition, carb. This time I rectified that. In essence, I had to change that fuel bowl body from an essentially 938SA to a 938SD, mostly by drilling out the air bleed from .025 to .031. Then I re-stamped the part with 938SD along with the other side that still shows 2756S. So now I'm working with what Carter seems to have concluded was their best version, a 938SD. The exception to the complete conversion is that I'm not using a check valve; I'm using a check ball and weight, which seems to work just fine in my application. I chose the ball design because I was able to complete the 60 second test with the ball, but NOT with the valve. (There are a few other items to mention later).

For now, I'll just get these following pictures out of the way.

user posted image

user posted image

user posted image

user posted image


Fuel Supply Circuit

This first picture shows an original OEM style fuel inlet valve and fuel flow shut-off needle assembly (bottom) and a second valve/needle (top) of a style often found in an acceptible rebuild kit, such as the Hygrade 518C. I say "acceptible" because the valve orifice is .088. An orifice greater than .090 is said to cause flooding problems for our Jeep applications.

user posted image

However, as you will see in further pictures, the original style is MUCH preferred IF YOU HAVE IT! I mean as long as it seals.

I found putting the valve to my lips (a short piece of correct diameter hose would have been easier) in the direction of normal fuel flow, while holding light finger-applied spring pressure against the needle assembly sealing the orifice in the valve, and blowing hard (I've been accused of being a blow-hard), resulted in a safe assumption that my valve is serviceable. And in final practice actually proved out, no flooding.

Note the original has a "straw" to suck through in the valve. That "straw" descends into the area of fuel from the fuel pump and likely results in less splashing inside the fuel bowl. It doesn't actually suck, but everyone knows what a straw does.

Also note the daintier (higher quality?) spring on the original as compared to the aftermarket.

Now on this next picture (valve and needle assembly all put together), note the height of the point where the float rests between the two styles. Use your eye to measure the height by starting at the threaded portions and move left to the spot where the float rests on the needle shoulder.

user posted image

The aftermarket unit is taller. The original style allows the float to settle better into an exact float adjustment (original on left).

user posted image

If you look very close, you'll see the aftermarket unit on the right side requires a float tab bent really badly, just to accomplish an acceptable, albeit just barely, float level!

As oldtime has implied elsewhere, the exact 5/16" float level setting is not so critical as is the correct orifice size and a valve that actually seals.

So those who must use an aftermarket valve, get the right sized orifice, make certain it seals, (viton/rubber tipped probably helps) and just do the best you can with the float level. The main thing is "have enough to feed the engine when needed, but not so much as to spill over and flood the engine". A worst case would be to try and get a little more bend to get it exact and break a tab off or something!

Two more pictures...

user posted image

user posted image

Float Drop: The 1.25" float drop setting is attained from two tabs on the float against the air horn body metal, so the aftermarket valve assembly isn't an issue in this adjustment.

user posted image

Accelerator Pump Circuit

There are several areas to address on this circuit. But before we can really understand them, one needs to really know the circuit and how it works. I'm going to try and help out there.

Before I got into this job I tended to think of the accelerator pump function in how it looks. You know what I mean, it has a long plunger handle at the top and it has a rubber doo-hikie thing on the bottom called a diaphragm, just like a toilet plunger!

You push down on it and sh!t pushes out. Clog fixed! Right? Wrong! (Theoretically a toilet plunger is pulled UP to fix a clog, but I ain't no plumber).

But that whole image is wrong! Wipe it from your mind. The accelerator pump in this carburetor (I can't say on other carbs, but I suspect it goes both ways), no on THIS carb, fuel is pushed into the carburetor throat when the diaphragm moves UP! Yes up. I know, that was news to me too. And boy was I embarrassed!

This picture helped me a lot though...

user posted image

But of course gas isn't green, just for this illustration.

See the white area below the diaphragm? That's manifold vacuum. That's the one that can be blocked (really messing up your circuit and causing a stumble) by an improperly fitted gasket! The lower gasket between the bottom body part (throttle valve housing) and the middle body part (the float bowl or Fuel Body).

Let me show you what I mean.

I had two gaskets, a new one (the passage in question has a small circle in pen written on the gasket material, bottom center)...

Note: a quick glance or cursory examination implies all is well. Upon closer inspection, mounting screws would re-center the gasket holes, blocking that passageway!

user posted image

Now here's the second on-hand gasket, slightly used, but apparently servicaeble as the Jeep runs good now. I used the second one (passageway in question now on the left side center).

user posted image

Much higher quality gasket. Where are these made? Where were they made? When were they made? You know what I mean. I could have cut a small amount out of the other gasket, but I was concerned for loss of integrity on the already narrow remaining gasket material.

I might add a thought here. If you get a pinhole in your diaphragm, you'll know something is wrong, because the fuel mixture will be improper as fuel is sucked from the pump housing (and fuel bowl) through the manifold vacuum source and into the combustion chamber.

In the diagram above, look close at the "Pump Jet" area, middle of the picture near the discharge weight and ball.

See how it looks like a nozzle? It is.

Now I might be wrong, but I think there is a major spray of fuel that comes from the accelerator pump circuit and goes out the carburetor's "main nozzle", which is that big one up in the middle of the carburetor throat at the very top, but apparently, there is also an amount of fuel that comes out of this "pump jet" shown in the picture, and EASY to find on your carburetor when you have it all apart.

When Carter started to upgrade the Carter YF, they ended up with their best final version - apparently - the 938SD (speaking just of the Jeep carbs). The earlier carbs with a check disc, and the later 938S, 938SA and 938SC with check balls were up-gradable to what was in essence the results found in a 938SD.

That upgrade (Service Manual, Paragraph F-27) involves drilling the jet bigger from .025 to .031 ONLY when changing to a 938SD - in essence, installing the metering rod 75-1500 and the main jet 120-166 that were in the upgrade kit. It also replaces the ball/weight/retainer clip with the check valve (apparently clip optional per oldtime). But the procedure doesn't appear to discuss making a new seat for the cone shaped check needle, as opposed to the previous upgrade to make a new seat with a steel ball and drift punch to fit a sphere aluminum check ball!

As I was going through my collection of steel balls, aluminum balls, c-clips and check valves, I read oldtime talking about the seat fitting a sphere like the check ball or a cone as found on the needle. I tried the leak-down (referred to as the 60 second test) for each configuration I had (several aluminum balls and two check needles).

I was finally able to make a successful "seat" with my steel balls (gotta love poetic license) and so I used a check ball design in my converted 938SD instead of a check needle and at least in my case, it is working fine. I couldn't figure out how to easily make a cone shaped seat in this conversion carb (which started out as the equivalent of a 938SA) and I couldn't get the needle to seat. I suspect a tiny common tool may well be an answer, but I envision a metal pencil point and some valve lapping compound.

Also, I drilled the pump jet to .0315 instead of .031, because it was the .8MM drill bit in the set I bought. Had to use my Dad's antique 1/4" drill to chuck it up because my newer drill chucks are too big.

user posted image

You can see the top of this check ball/valve area in this next picture. It is the hole which has the tiny c-clip holding in the ball and weight at the lower left.

user posted image

In that picture you can see the drilled orifice on the far wall where it goes into the carburetor throat.

So I think an accelerator pump circuit summary might be:
Careful of your gaskets, do the new seat correctly and verify by getting a successful 60 second test! Drill the pump jet to .013 if you have the metering rod 75-1500 and main jet 120-166, otherwise, if you're using the older carb's (938S, SA, SC) rods and jets leave the pump jet orifice at .025. Oh and don't twist the diaphragm rubber with the mounting screws.

Earlier I said I would mention why I think many conversions failed. I'll say that in three words, "Shade Tree Mechanics". What I have done here (pulling a drill gun and firing away at the orifice), is typical of shade-tree mechanics. We are all different. And we will all jump into a job with differing levels of gusto. Why should we think the past owners were any different? Do you think they all had the right tools on hand to drift punch the new seat, to drill the new hole with a drill bit they didn't own, to strike the hammer at the right amount of force, etc. These are precise things and not everyone will do the job right, or completely.

Last thought, almost ANY new accelerator pump in a kit would be a successful replacement. If you're re-using your old pump, at best store the diaphragm in gas. At worst, reconstitute it in gas and then verify the rubber feels and looks good by a bit of finger manipulation. And why not keep the old one if it is okay? Next time you do carburetor work the spare parts bin looks more impressive.

QUOTE: "That is to say a poorly fitting gasket between the fuel bowl body and the throttle valve body, might very well shut down your vacuum assisted portion of the accelerator pump design. Would that cause a stumble?"

Oldtime: "Exactly true. Dave provides us with accurate information concerning various forms of YF throttle body gaskets. I have also noted all these exact same differences between OEM and Hygrade 518C inlet valves which Dave has detailed precisely. I fully agree that the Hygrade inlet valve is serviceable yet overall it is inferior. The biggest single concern is that it drastically changes the float setting. Just a reminder to always keep your original (superior quality) parts no matter what. Even the original inlet needle assembly can be rebuilt if you take your time to do it right. Thanks Dave for this excellent contribution to our Carter YF database."

Dave: Alright oldtime, thanks for the help! Also, in the course of my learning after oldtime brought my awareness to the fact there are TWO procedures (Service Bulletin 1951 and a later Service Manual Procedure circa 1960 and later) that I had confused into ONE DOCUMENT, I located the following -- A complete online scanned PDF file of the Universal Jeep Service Manual, 281 pages and dated 1965, at Willys Manuals. Right click the link on that page to download and save it to your computer while it's still available.

Manifold Vacuum and Throttle Valve Body Mounting to Intake Manifold

While still on the subject of the accelerator pump and it's dependence on the manifold vacuum - spoken of earlier in the gasket alignment comments - I decided to discuss this next issue here - mounting the carburetor to the intake manifold.

As we all probably know, there's a 1.5" bore on these carburetors. It stands to reason that the manifold is a 1.5" bore, and the mating gaskets are a 1.5" bore. However, I found some differences in such assumptions.

First of all, gaskets. Look at these two gaskets side by side. Note how the left gasket is larger overall.

user posted image

It appears I failed to get a picture of the larger gasket on the flange, but this following picture shows the smaller gasket.

user posted image

Even the SMALLER gasket has a larger bore (not that important), but at least the holes line up well. The larger gasket had a poor fit on the manifold vacuum hole, and was pretty bad on the bolt holes too. The larger gasket was also getting pretty darn skimpy on remaining material around the bore to seal overall. I went with the smaller gasket.

After test driving, I eventually checked my seal (and the throttle shaft wear) by spraying bursts of starting fluid (ether?) around these areas. The gasket seal (and shaft wear) were acceptable, as I didn't get an increase in engine RPM when I sprayed the areas. But, I seem to remember reading somewhere that I should have tested with engine idle set as low as possible, and perhaps with it hooked up to a tachometer. That would make the test results more easily detectable than simply listening at normal 700 RPM idle speed.

So the gasket thing is the first item.

The second item to speak of is the alignment of the bore! I found my throttle flange (from both carbs) wasn't exactly in line with the intake manifold bore. On the rearward side of the manifold bore there was about an 1/8" ledge that induction air/fuel mixture runs smack into on it's downward flow. That's counterproductive to good induction and that translates directly to engine performance. If these engines are already poor performers, it stands to reason we should try and improve flow when we can.

I already had a clean rag stuffed into the intake manifold and I left it there for my "porting" procedure. Next I hooked up the shop vac to suck right near my work area. Then I got the dremel tool out with a suitable grinder bit and I carefully removed the 1/8" shelf metal on the manifold that stuck out into the flange bore. It didn't take much, and I very slightly knocked down the sharp edge of the entire bore with a slight bevel grind. Thereafter the bore had a nice clean visible drop into the manifold. And metal filings were easily removed by the shop vac and carefully pulling the rag from the manifold.

Oldtime: "Very well done Dave, and your pictures are a big help to many.

"I too have noticed all these things that you have mentioned. I fully agree with your assessments. We eventually come to accept that there seems to be an never ending amount of detail to observe. The observations of detail are seemingly infinite. Things like fine tuning the carburetor via exhaust audio. Or calculations concerning optimum fuel characteristics. I dare say our full understanding of the Carter YF 938 series will not be ending any time soon. After all we have not even touched upon the fuel conditions as they compare with rocket science. Pressures, burn speed, flash points, burn temperatures, CFM, flow characteristics and so on. Yet in the meantime this post will help us to get better performance from the Willys Hurricane."

Continue to A Tale of Two Carbs, Part 2 on CJ3B.info.

Thanks to Dave and Oldtime. -- Derek Redmond

See also YF Carburetor Tips and the Carter YF Carburetor Documents on CJ3B.info.

Return to CJ-3B Specs and Tech Tips.

FacebookVisit CJ3B.info on Facebook.

CJ3B Home | Contents | Search | Links | 3A and 3B Community

Last updated 10 April 2016 by Derek Redmond redmond@cj3b.info
All content not credited and previously copyright, is copyright Derek Redmond