JNaws Carburetor Adventures

JNaw

XS650 Enthusiast
Messages
73
Reaction score
38
Points
18
Location
Calgary, Alberta, Canada
Hey all, got the bike up and running and been enjoying the good weather this year so far. Still tinkering with the jetting as I can't get it quite dialed in, just want to share what's going on and brainstorm with you guys

79 xs650 special (non special gastank, dual fed peacocks/carbs/non vacuum, new carb holders)

Jet specs: (3428ft elevation),
145 main
Middle clip on needle (3 from bottom)
32.5 pilots (too big I think but working on it)
Passes the diaphragm test.
Canadian spec bike
Synced best as I can as I don't have access to a vacuum source, just used the "hands over the pipes", as well as dead cylinder method.

Mods:
Uni pods
Emgo tapered mufflers
New ignition coils/plugs/battery/stator (plugs gapped correctly last year)

1. Engines bogging/cutting out/running out of fuel at highway full throttle runs up hill, opening gas cap seems to help/mostly eliminate it. Is there a procedure to enlarging the venting hole in the gas cap?

2. Ties in with 1, full throttle runs there is a minor surge (not the bogging but just a slight power surge) Im having a hard time thinking it would be lean with 145's, could this be a symptom of too rich? (I'm going to try 142.5 mains if I can't get my hands on them, as bike seems to have slightly more power when cold, but very minor difference)

3. I've got 2.5 turns out mixture screw on one carb, and 1 turn out on the other, with the same jets, no noticeable vacuum leak with spraying brake clean all over the carbs while running, what gives?

Im going to try cleaning the carbs and filters again and see if that changes anything, but sort of stuck in my thoughts here. I have only tuned slide carbs so not super familiar with CV, although working through the tuning guide for CV's when Ive got time.

Let me know your thoughts and what I should try. Thanks all.
 
Last edited:
How do the spark plugs look? That should give a clue.
Will take a look but last I looked they were slightly lean with the 140's.

@650Skull, took the gas cap apart last year and I cleaned it best I could, is this a yearly maintenance item due to the ethanol in the gas breaking down the rubbers faster than expected? I've even replaced the seals with new ones and already starting to crack...

Thank you very much for the link, I couldn't find that in the tech section but maybe I wasnt looking hard enough. Ill definitely take a read through!
 
Yes, you need to clean that gas cap vent again before you can do any tuning. No, I don't think it's a common or reoccurring issue. I've never had it happen in many years of ownership. I think you probably just didn't get it cleared out good enough the first time. Yes, I think 145 mains are too big, and the only reason that you're able to run them is probably because that plugged cap vent is causing the bike to run lean, starving it for fuel.

It's very easy to over-jet CV carbs because they work on engine demand, only feeding it what it can use. That's why if the mains are too big, the bike may still seem to run fairly well. When you whack the throttle wide open, the slide doesn't necessarily open all the way, it only lifts as far as the engine can use. Those too large mains don't flow at their full capacity.

Now let's talk a little about the 650 CV carb history. Here's a chart I made up listing the stock jetting for the various years and you'll notice that the '78-'79 carb set came with the largest mains used in any 650 carb set .....

Carb Specs.jpg


But in order to run those large mains, they had to reduce the needle jet size (from a Z-8 to a Z-2). However, the upper midrange on the '78-'79 carb set is still on the verge of being too rich. This means that any increase in main jet size, even just one, normally pushes that upper midrange over the edge and makes it too rich. This causes stumbling and break-up in the upper midrange, say in the 4 to 5K range, under heavy or full throttle applications. The "fix" is to lean the needles a step, and as I said, this normally needs to be done on the '78-'79 carb set with even just one size up on the mains. You have to realize that the 3 circuits in you carb overlap each other slightly. Changing one will affect a portion of the one next to it .....

Carb_Circuits.jpg


So, you increase the main jet size which makes the upper midrange too rich due to this overlap thing. Then you lean the needle a step to fix that, but that also leans the upper part of the pilot circuit, and that may require a larger pilot to "fix" that. The stock pilots may be fine - until you lean the needles. Also, Mikuni needle slots are counted from the top down .....

Needle Slots.jpg


..... so to lean the needle a step, you move the clip from the stock #3 slot up one to the #2 slot.

When tuning CV carbs, you can't really use throttle opening amounts to determine what area or circuit needs changing like you do on normal slide carbs, you must rely more on the RPMs that the trouble occurs at. Also, for testing, you must work the carb hard. CV carbs are very forgiving and will mask over small jetting "glitches" unless pushed hard. So, to test the midrange to main transition after installing larger mains, from about 3.5K in 2nd or 3rd gear, give it full throttle and run up to around 5.5K. Look for any stumbling or break-up in about the 4 to 5K range. If you get any, the needle will need to be leaned. To test the pilot jet size and pilot circuit to midrange transition, find a large empty parking lot you can cruise around in in 1st gear at about 2500 to 3K RPMs. Check to see if the bike runs smoothly. If it surges, you're lean, stumbles or breaks up, you're probably too rich. Then grab a big handful of throttle and run up to about 4K. If you get a big flat spot or hesitation, you're lean, break-up or stumbling, you're too rich. Now I know you won't normally be riding the bike like this (full throttle applications) but like I said, you need to work these CV carbs hard when sorting the jetting or you won't find the flaws.

When I re-jet CV carbs, I like to work from the top down because the mains influence the needle setting which, in turn, influences the pilot jet size choice. For '78-'79 carb sets on bikes with the typical "pods and pipes" mod (like yours), I've found 140 mains, 30 pilots, and the needles leaned one step works well usually. I was able to use 142.5 mains in a set but that was only because I installed leaner 5Z1 needles from an XS400.
 
Last edited:
Yes, you need to clean that gas cap vent again before you can do any tuning. No, I don't think it's a common or reoccurring issue. I've never had it happen in many years of ownership. I think you probably just didn't get it cleared out good enough the first time. Yes, I think 145 mains are too big, and the only reason that you're able to run them is probably because that plugged cap vent is causing the bike to run lean, starving it for fuel.

It's very easy to over-jet CV carbs because they work on engine demand, only feeding it what it can use. That's why if the mains are too big, the bike may still seem to run fairly well. When you whack the throttle wide open, the slide doesn't necessarily open all the way, it only lifts as far as the engine can use. Those too large mains don't flow at their full capacity.

Now let's talk a little about the 650 CV carb history. Here's a chart I made up listing the stock jetting for the various years and you'll notice that the '78-'79 carb set came with the largest mains used in any 650 carb set .....

View attachment 325636

But in order to run those large mains, they had to reduce the needle jet size (from a Z-8 to a Z-2). However, the upper midrange on the '78-'79 carb set is still on the verge of being too rich. This means that any increase in main jet size, even just one, normally pushes that upper midrange over the edge and makes it too rich. This causes stumbling and break-up in the upper midrange, say in the 4 to 5K range, under heavy or full throttle applications. The "fix" is to lean the needles a step, and as I said, this normally needs to be done on the '78-'79 carb set with even just one size up on the mains. You have to realize that the 3 circuits in you carb overlap each other slightly. Changing one will affect a portion of the one next to it .....

View attachment 325637

So, you increase the main jet size which makes the upper midrange too rich due to this overlap thing. Then you lean the needle a step to fix that, but that also leans the upper part of the pilot circuit, and that may require a larger pilot to "fix" that. The stock pilots may be fine - until you lean the needles. Also, Mikuni needle slots are counted from the top down .....

View attachment 325638

..... so to lean the needle a step, you move the clip from the stock #3 slot up one to the #2 slot.

When tuning CV carbs, you can't really use throttle opening amounts to determine what area or circuit needs changing like you do on normal slide carbs, you must rely more on the RPMs that the trouble occurs at. Also, for testing, you must work the carb hard. CV carbs are very forgiving and will mask over small jetting "glitches" unless pushed hard. So, to test the midrange to main transition after installing larger mains, from about 3.5K in 2nd or 3rd gear, give it full throttle and run up to around 5.5K. Look for any stumbling or break-up in about the 4 to 5K range. If you get any, the needle will need to be leaned. To test the pilot jet size and pilot circuit to midrange transition, find a large empty parking lot you can cruise around in in 1st gear at about 2500 to 3K RPMs. Check to see if the bike runs smoothly. If it surges, you're lean, stumbles or breaks up, you're probably too rich. Then grab a big handful of throttle and run up to about 4K. If you get a big flat spot or hesitation, you're lean, break-up or stumbling, you're too rich. Now I know you won't normally be riding the bike like this (full throttle applications) but like I said, you need to work these CV carbs hard when sorting the jetting or you won't find the flaws.

When I re-jet CV carbs, I like to work from the top down because the mains influence the needle setting which, in turn, influences the pilot jet size choice. For '78-'79 carb sets on bikes with the typical "pods and pipes" mod (like yours), I've found 140 mains, 30 pilots, and the needles leaned one step works well usually. I was able to use 142.5 mains in a set but that was only because I installed leaner 5Z1 needles from an XS400.
Really appreciate this thank you, I now realise it's rpm based not throttle based which will help alot with the tuning.

When I originally had 140's in it, it sounded harsh and had no power at top end. 145's solved that problem but had the stumble at 4-5k as mentioned. Leaning the needle helped fix that but I still am too rich on the main (the supplier sent 145 instead of 142.5 as they weren't in stock)

But now that you mention the vent could be causing the whole system to be lean, that may be the culprit.

As for the float height, the bs38's that I have, have the plastic floats with a plastic tang, is there a trick to adjust the height or am I looking at swapping to a brass set?

I'll try the 142.5 and let you know how that goes, gotta order and get it shipped still.
 
I never saw plastic floats in BS38s. I wouldn't know what to set them to by measuring. I think the only way to set the fuel level correctly would be to use the clear tube method .....

SightTube6.jpg


Correct fuel level would be at about the seam between the bowl and carb body.

If the floats and setting them gives you problems then yes, I would switch back to original floats. I've discovered that the floats used in the BS38s are nothing special, so no need to pay big bucks for them from a specialty 650 place. They're nothing more than the commonly available Mikuni VM24-171, and retail for about $15 .....

Mikuni-VM24-171-Genuine-Brass-Float-Dimensions.jpg


Compared11.jpg


Compared12.jpg


Installed2.jpg


Installed3.jpg
 
Set carb float height to 24mm (one of the floats was full of gas, got brand new brass floats installed), 30 pilots and 142.5 mains, ultrasoniced the carbs and bike runs better than ever.

It seems to like the pilot jet screw set all the way in or half a turn out though. May have to go back down to a 27.5. surprised it ever ran on a 35, must've had a plugged pilot circuit.

Does the mixture screw meter air or fuel on these carbs?
 
Actually, it meters a portion (roughly 25%) of the air/fuel mix coming from the pilot jet. It is a true "mixture" screw, not an air screw. Turning it out makes the idle circuit richer, turning it in makes it leaner.

It's odd that your mix screw needs to be almost closed. The factory spec for your carb set is 2.25 turns out, and it usually likes that setting, often even a bit more (2 1/2 to 2 5/8 turns out). There's an air jet located along the outer edge of the bell mouth that feeds the air to the pilot circuit. Looking at the bell mouth, it would be the one on the lower right. Make sure it's clear .....

LabeledCarbMouthsSmall.jpg
 
Actually, it meters a portion (roughly 25%) of the air/fuel mix coming from the pilot jet. It is a true "mixture" screw, not an air screw. Turning it out makes the idle circuit richer, turning it in makes it leaner.

It's odd that your mix screw needs to be almost closed. The factory spec for your carb set is 2.25 turns out, and it usually likes that setting, often even a bit more (2 1/2 to 2 5/8 turns out). There's an air jet located along the outer edge of the bell mouth that feeds the air to the pilot circuit. Looking at the bell mouth, it would be the one on the lower right. Make sure it's clear .....

View attachment 327672
Odd even with a #30 pilot? Stock is 27.5 I believe. Having it in would tell me I'd need a smaller pilot jet yeah?

I'll give that hole a poke again, blew air through it though and it seemed clear.
 
Yes, even with a 30 pilot. The screw may dial in a little less, like maybe 1/4 to 1/2 turn less, but it wouldn't (or shouldn't) need to be closed almost completely.
 
Yes, even with a 30 pilot. The screw may dial in a little less, like maybe 1/4 to 1/2 turn less, but it wouldn't (or shouldn't) need to be closed almost completely.

Okay will check tomorrow on those holes.

Sent a pm but thought I'd mention here as well.

When engine is at operating temp, it's having a hanging idle after blipping throttle as well. This doesn't occur at startup


These two things are contradicting, the hanging idle being lean and the pilot screw all the way in being rich?
I recall having to go up to 35 pilots to remedy the hanging idle, but the pilot screw was operating correctly before the whole ultrasonic clean. (2-3 turns out on 35's) It's definitely happier with 30's though

Anyhow, even with the 142.5 mains and needle at the same height (middle slot) it's having the loss of/low fuel feeling after extended 3/4 throttle to full throttle runs on highway (Although full power runs are definitely stronger with the 142.5) letting off throttle for abit and then returning remedies for a bit before coming back.

I'm wondering if the pet cocks aren't flowing enough, although it's the older style tank with 2 petcocks so I find that hard to believe? No inline filter currently as I thought that was the issue. I did notice the petcock fuel nipples have a mesh screen on them, starting to wonder if those are restricting flow. I did rebuild the petcocks and cleared them out when I first got the bike (completely full of aluminum oxide)

The fuel cap is venting properly, and I even added extra holes around the rim of the ventcap (it's a 74/75 style lid) where the air flows to hopefully vent better. To no avail. She's squeaky clean

Any ideas? It's got a new rotor, battery, and relatively new ignition coils too.
 
Last edited:
Also, would the needle jet to main transition have issues that don't occur instantly? Because the "leaning" feeling is AT 5k rpm. But not all the time.

Edit: maybe a sticky float too? Didn't bother to check if the float holder pin holes are oversized.

I'll try lowering the needle another slot and see if that does anything I suppose, worth a shot.

Just getting annoying taking the tank and carbs off everyday haha
 
Last edited:
Tore carbs apart and made sure everything was clear. (Pilot pickup tube was sort of clogged)

Noticed carbs have 4n8 needles. These are definitely 78/79 style carbs. Seems like new needles are in order. Already have z-2 needle jets

Anything else need to be changed with the needles?

Edit: carbs now like around 1 1/2 turns out on the 30 pilots. Right side revs engine up when all the way in but still stays happy about 1 turn out.

Rev matches are crisp now, but still sometimes has a slight hanging idle (worse when fuel is turned off, will stay around 2k rpm, fuel on will hang for a sec and then drop). May have to look for air leaks again or the ignition advance system, or raise fuel level in carb? Althought when I checked the advance last time she wasn't sticking.
 
Last edited:
Well, I'm surprised the bike ran as well as it did if it has 4N8 needles in it. They're a full 5mm shorter than the correct 5O2 needle, which would make them very, very rich .....

BS38Needles.jpg


The taper and diameter of the tip are pretty much the same .....

BS38Needles3.jpg


BS38Needles4.jpg


..... but being so much shorter is a real issue. Get some correct 5O2 needles and I bet your carb problems will be solved.
 
Well, I'm surprised the bike ran as well as it did if it has 4N8 needles in it. They're a full 5mm shorter than the correct 5O2 needle, which would make them very, very rich .....

View attachment 327840

The taper and diameter of the tip are pretty much the same .....

View attachment 327841

View attachment 327842

..... but being so much shorter is a real issue. Get some correct 5O2 needles and I bet your carb problems will be solved.
That's what I'm thinking too. Bike runs amazingly well besides the highway haha.

New needles ordered. Will report back when I've changed them out.
 
Last edited:
Make sure you assemble the needle correctly in the slide .....

'78-'79 Slide.jpg


The needle should be spring-loaded from the bottom. I'm thinking maybe it was assembled with the little spring on top. That would make the needle sit much lower and could explain how you're able to run that much shorter 4N8.
 
Make sure you assemble the needle correctly in the slide .....

View attachment 327877

The needle should be spring-loaded from the bottom. I'm thinking maybe it was assembled with the little spring on top. That would make the needle sit much lower and could explain how you're able to run that much shorter 4N8.
I wish I took photos but there is no spring. Darn.

Its a brass slide, with a bore in the middle where the needle sits down, with a plastic thick washer with four holes on it that covers over the top of the jet, then the spring and lid go over top of that.

Is it possible it's got an older style slide in it and I'm looking at a whole replacement slide? Here's a photo for reference on what I've got. Or at this point would switching to the appropriate needle jet work fine? I'm not sure on interchangability on the slides between carb revisions?

Just starting to see how Frankenstein this is, no idea how I didn't catch this the last 6 times I tore it apart, didn't know better I suppose since it ran pretty alright.
Screenshot_20240606_130621_Chrome.jpg
 
Last edited:
I wish I took photos but there is no spring. Darn.

Its a brass slide, with a bore in the middle where the needle sits down, with a plastic thick washer with four holes on it that covers over the top of the jet, then the spring and lid go over top of that.

Where can I get this spring? Would a pen spring work if I cut it shorter?

Just taking a quick look, is it possible it's got an older style slide in it and I'm looking at a whole replacement slide? Here's a photo for reference on what I've got.

Just starting to see how Frankenstein this is, no idea how I didn't catch this the last 6 times I tore it apart, didn't know better I suppose since it ran pretty alright.View attachment 327886
That looks like the slide/needle assembly in my 77 XS650D. Different to a 78/79 assembly.
 
That looks like the slide/needle assembly in my 77 XS650D. Different to a 78/79 assembly.
Yep just checked part numbers on the parts diagram and it's a 75 year assembly. Different part #

75: 256-14940-00-00
79: 2F0-14940-00-00

My next question is how can I make this work, or am I better off sourcing a 79 complete assembly.

Bah.

edit: seems heiden tuning has replacements: https://www.heidentuning.com/xs650-...sembly-for-late-mikuni-bs38-carbs-detail.html
 
Back
Top