Jeff20B

iTrader: (13)

I had forgotten how tedious and time consuming they are to cut out. But as usual to make things efficient I did a production run and made four. It used up the whole sheet. You can tell I don't do this very often.

Jeff20B

iTrader: (13)

I had to use some of that carb dip stuff today. Very nasty, but it got rid of the gum and varnish!

Jeff20B

iTrader: (13)

I now have four carbs stripped, torn down, cleaned, all four boosters and primary venturis removed. Ready for hogging and boost prepping.

I'll follow this guide I just typed a couple days ago. You can too! https://www.rx7club.com/1st-generati...ealed-1083467/

Jeff20B

iTrader: (13)

I have new data to add to the list.

I tested the 25.4mm carb after I updated its venturi profile (basically hogged it out under the narrow part and kept the taper as accurate and straight as possible compared to the slightly hourglassed shape from before).

I swapped in a set of 60 main primary air bleeds because I figured since more air can now flow through, it might want more fuel.

I was wrong.

It ran too rich at part throttle cruise, around 11, and was herky jerky and just didn't run all that well over all.

This carb never ran this bad before. So I swapped the stock 70s back in and everything was fine. Carb seems to have a good amount of power and drivability. I'm happy with it.

But how were the secondaries? I think I finally found the threshold that would get me down past the 10.0 limit of the wideband. It was a set of old school (small funnel hitachi) jets I drilled to 131. This caused the wideband to read 10.0 as soon as you open the secondaries, then it would creep to 11, then 12, then 13! Gasp! Why am I running lean? Could it be from the small funnel design of these early jets? Maybe. Is 131 just too small?

So to test one theory, I swapped in a set of known working modern Nikki deep funnel jets drilled to 137. These ran at 10.0 in the other carb every time I went into boost without fail. This will test whether this carb tends to run lean in boost for whatever reason. If it behaves the same as the other carb, ie 10.0 all the time, problem solved. Just drill another set of deep funnel jets to 131 and try again. If, however, the deep 137s go lean like the shallow 131s did, then this carb has a fuel delivery problem. Let's hope it's not that.

I'll have to test this carb at some point soon and report back.

Jeff20B

iTrader: (13)

Ok I have an update. At long last I finally got one of these carbs tuned 100%! It only took a little over a year.

The first thing I did was to test a theory mentioned above where the shallow funnel shape of the early Hitachi jets might be hurting fuel flow. I was right. I took a set of modern Nikki jets and drilled them out to 131, same as the hitachi jets. Or it was closer to 130. I'm just going by what the digital caliper says about the micro drillbits.

The result was secondaries that would only go to 12.9 on the wideband. Ok, at least it's not 13, but too lean for comfort.

But something else I did on the carb should have produced a leaner mixture as well, proving the deeper funnel shape of the modern Nikki jets works better. I swapped from the drilled 122, or more like 123 primary fuel jets to some drilled 120s. These are actually 1.20mm and are as accurate as the digital caliper says they are. lol

Anyway it got my primary AFR from a somewhat rich 11 at part throttle cruise to more wankel=awesome territory of 13. This means better gas mileage. We all want that, right?

So seeing as my primaries are this much leaner, but my AFRs in boost are slightly richer than 13 with secondaries the same size as before but with a much better shape (deep funnel) this time, 12.9 in boost was a step in the right direction. But still grossly too lean.

So now that the primary circuit was tuned as well as you'd expect, all I needed to do was focus in on the secondaries and get the AFR into the 11s and keep it there.

So I went with my original idea of using the 137 or rather 136 (yep) drilled jets in the secondaries.

It worked!

I saw 11.something the whole time my foot was planted. Now that's what I wanted to see! No more pesky 10.0 where you have no idea just how rich it is, and no more worry about leaning out the longer you kept your foot in it like the shallow hitachi jets did.

For those wondering, carb specs are:

25.4mm venturis
118 primary short slow bleeds
stock 46 primary long slow bleeds
pri air 70
pri fuel 120 (drilled)
sec air 91 (stock fuel jets in modded air bleed head)
sec fuel 136 (drilled)
sec short and long solder filled

Engine is a 74 ported R5 13B
S4 NA rotating assembly
GSL-SE spec exhaust ports
rebuilt S5 turbo
full 3" exhaust (it creeps to about 7psi)

Fuel system is an MSD 2225 fuel pump (inline/external style and is very quiet on rubber standoffs)
Mallory 4309 fpr
5/16" send line (stock)
3/8" return line (just some steel brake line from Napa bent to fit under the car)
GSL-SE gas tank
12V relay which makes life easier when getting a fresh carb up and running

Any of you 12A guys can copy this setup, just with a 12A. I'm going to look at building a 12A turbo based on all of this as soon as I can.

Jeff20B

iTrader: (13)

I saw 12.9 on the wideband with this carb again. How? Why? I'll fill you in while it's still fresh in my mind.

As I was testing carbs today, I kept getting the annoying secondary siphon effect and it was leading to all kinds of flooding problems every time I'd floor it for like one second or long enough to hear the BOV sound. It drains the float bowls faster than they can be filled at 2.5psi which pools gas on the secondary butterflies while leaning out the primary circuit. The carb's idle circuit starves while the secondary circuit is fed more than it can take. This effect can last a really long time, up to 1/8th mile or more, or it will continue as long as the engine is running and providing a vacuum and a running fuel pump. And even when it stalls or you shut it off, there is still a bunch of gas sitting on the secondaries making restarting difficult.

wat do?

Apparently I never experienced this type of flooding until I started doing solder filled secondary slow air bleeds in all my carbs. Then some carbs would flood more than others leading to all kinds of head scratching due to inconsistant results.

The reason I started doing solder filled sec slows was to help reduce the delay or hiccup or bog that everyone gets every time they do a boost prepped Nikki and open the secondaries, due to there being only one accel pump. On my car it was originally about 1.5 seconds. That is an eternity to wait for anything to happen when you've floored it. It was unacceptable. So I did a bunch of research and filled some air bleeds with solder, totally blocking them off. It reduced the secondary delay way down to 1/4 second average wait time. Then I modded the main air bleeds for small fuel jets (92) and it reduced it down further.

All the carbs suddenly had a tendency to want to flood more easily but because it only happened to one or two carbs more severly out of the four or more I was testing, it slipped under the radar but I can clearly remember one carb complaining through flooding (secondary siphon effect) right after I installed the solder filled slows. I was doing so much back to back testing it never got resolved.

Fast forward to today. The same easily flooded carb was being tested and it flooded itself out faster than the bowls could fill at 2.5psi, as mentioned above. I shut it off and removed the carb hat. Sure enough, two bodies of water, I mean gas in the secondaries while both bowls were at the bottom of the windows. Hmm... Could it be... nevermind.

So I threw the 25.4mm carb on, its specs mentioned one post above this one, and tried some slightly bigger primary fuel jets at 124 this time. Also swapped the top I had been testing on the easily flooded carb onto this better known carb. If it floods this carb, I'll know it's the needles and seats. Or so I thought.

Yes it flooded this carb just as easily as the easily floodable carb. What? Why? How? And why did it only happen when I'd floor it for one second in 1st gear but not when I'd break them loose in 3rd gear in warm weather? Why would the secondary siphon effect behave in such a way? It must be the boost basically preventing the siphon from getting started, however in vacuum or at low RPM such as in 1st gear when you want to just rev it up to hear the BOV or generally show off, it creates the siphon effect and nothing can stop it, other than to drive it for 1/8th mile or more, or let it stall and hope it will start back up.

Well anyway I simply swapped in a set of nickel plated 60s and there is no more secondary siphon effect! I can floor it for a second and it happily returns to idle. That's the good news. The bad news is the secondary delay is slightly longer now and the tune it slightly leaner.

See, I was relying on the secondary siphon effect before without even knowing it! This is how I got away with such tiny fuel jets! A nominal tune of around 11 in boost with 137 jets was normal. Now I'm seeing 12.9 on the wideband at the top of 3rd gear with the nickel plated 60s reinstalled so I must up my jets somewhere between current 137 and the stock 160s. I don't think I'll need to go as big as 160s, hopefully. I probably won't even need to go to 150s. I'm thinking I'll try 140 next and if that's still too small I'll try 145. Or maybe I'll do it the j9fd3s way and start with the 150s and work my way down until I get back to 11.

I'll just have to put up with the secondary delay, I guess. Or until the next breakthrough. It's not that long of a delay though - I'm not offended by it or anything. And Jingkun's carb at 22mm didn't have one at all. However his carb couldn't break them loose in 3rd gear either. I guess I traded a little drivability for power. Perhaps I've reached the edge of how far we should hog our venturis. If we need more power, bad83 proved that a Nikki, with stock primariy venturis, can make 300HP and break a Y rear iron. I think if I had to do it again, I'd have gone with smaller venturis, but I do love the better transient response I get here at 25mm.

Your thoughts?

Jeff20B

iTrader: (13)

This next bit is interesting. I tried drilling out a set of my solder filled slow air bleeds with the smallest micro drill bit I currently have at .35mm.

First let me back up. I swapped the 25.6mm carb in. The big one that requires a bit of attention to drive it (until you get used to it). The primary tune is still a little rich as I don't have a set of 126 jets yet.

Anyway, the secondary air bleeds were swapped for a set of solder filled and drilled jets at .70mm. These are the smallest I've ever tried in the secondary main air bleeds. Yes, primary and secondary air bleeds the same size! Did it work?

Hang on, I also have to tell you I swapped in a set of nickel plated 60s because I discovered they reduce or eliminate flooding issues that I've had ever since I swapped to them last year. But won 't that increase the secondary delay? Due to 60 numbers of additional (unwanted) air?

Yes. The secondary delay was lengthened to about 1/8th of a second. Kinda annoying.

But how did it drive otherwise?

Not as good as you'd think. When in boost, after the delay, it was glitchy. The same kind of glitchy I get when I swap stock 60 primary air bleeds into a carb that prefers 70s. This means the air bleeds are too small.

Cool. I reached the lower limit.

So primary air bleeds need to be bigger than 60 but smaller than 90, and the secondary air bleeds need to be bigger than 70 but smaller than 100.

How were the AFRs? I saw 10.5 with 140 secondary fuel jets. So not bad. Keep in mind my primary tune is still a bit rich.

Getting back to the .35mm drilled short slows. I swapped them in and determined the 70 air bleeds were the reason for the glitchiness so I swapped the 91s back in. The result was somewhat mixed because I did some mixed driving, but in short, this is a better combo.

The secondary delay is shorter (10th of a second now or sometimes even shorter) and the secondaries aren't glitchy. The AFRs are kind of hard to remember because I saw 10.0 and 14 at different times. I also tried pulling the e-brake to add more load to the turbo as I floored it at low speed going up my test hill. Pretty interesting.

Anyway after all that the carb never once flooded out! I'm a happy camper.

The trick is to allow just a smidgen of air in through the channel normally occupied by the nickel plated 60s. This skoshi of air is required to break the siphon effect and has no other purpose in a boost prepped carb with mechanical secondaries. So the "trick" is to make it as small as possible so it doesn't add much to the air bleed "numbers" total. For example I currently have 91 + 35 = 126 total "numbers", for lack of a better term, of secondary air. I'd like it to be smaller because it will reduce the secondary delay to probably zero*, but it may cause other issues.

*The last time I got rid of the secondary delay with any sort of reliability was when it had the 91 air bleeds and the short slows were filled with solder. So that means just 91 total numbers. I could floor it at any time and it wouldn't delay, but it also would flood out any time you'd floor it and let off quickly ie during testing. Not very fun.

My current thinking is to try maybe 80 air bleeds while leaving the slows at 35. 70s were too small. 91s work with decent AFR but still a little delay. Thoughts?

Jeff20B

iTrader: (13)

Update on the 26.5mm carb.

So I drilled out a set of old stock Nikki jets to 124 and made sure they were not snug across the widest part of the flutes. Yes these cheap Chinese drill bits are quite inconsistent. Some areas are 1.23mm while others are 1.24mm. It is supposed to be a 1.25mm bit. Oh well.

I removed the old factory 130 Hitachi jets which were closer to about 1.28mm which in this warm weather is way too rich. I would see high 11s to low 12s at part throttle cruise. Terrible gas mileage.

The new generously drilled 124s gave a more wankel=awesome friendly 13.0 or so at part throttle, and they didn't change the behavior of the carb for the worse. If my Nikki sense is right, I think 124 will be fine for warm weather and 126 will be fine for cool weather. Thoughts?

As for the secondary air bleeds, I swapped in a set of drilled 81s to replace the factory 91s to check for glitches like the drilled 73s produced.

Lastly the sec short slow bleeds are still the drilled 35s.

Time to take it for a test drive.

The result was good primary driveability. Good primary AFRs in this warm weather. I think the primary tune is set now.

The secondary still had the annoying delay, even after all I've done!

But wait! There is an inconsistency that I'll explain.

I noticed something rather interesting and welcome that came out of nowhere. I can floor it at really low RPM and there is no secondary delay or bog at all! Yes, even with mechanical secondaries, there is no bog from the lowest RPM I tried which was 2500 or 3000 up to like 4k or so. In 2nd and 3rd, it didn't matter. It would just sound like the engine gets louder. No flooding, no real complaints at all.

However when I'd wait to punch it until after 5k, there would be a secondary delay every time.

Have I been driving this thing all wrong this whole time? I mean who waits until 5k or 6 grand to floor it so the secondaries open anyway? Besides me, who else does this?

Think about it. This carb is just fine if you floor it at say 3000 and just leave your foot planted. The result is the intake noise gets louder instantly, there is no bog or hiccup or delay at all, and if you keep your foot planted, it will spool the turbo up and the wheels will break loose in 2nd or 3rd gear on dry road before 8k where I shift. Doesn't matter which gear you're in.

Also the AFRs are back in the healthy range of 11.something. Not glitchy at all.

Then you let off and the BOV makes its loud pshhh and the wideband goes back to its healthy idle numbers of around 12.2 or so. Sometimes 11.6, then creeps back to 12. The engine doesn't run funny afterward or anything. I think it likes this kind of driving!

I think I solved it, guys. Finally. After a little over a year of dealing with the constant companion, it turns out at least part of the issue was the way I was driving it. I need to just lay into it a little sooner and keep it there from a slightly lower RPM and let it climb like on a dyno or something. Then everything will be fine.

Gee, isn't that how it was supposed to be this whole time anyway? Smack myself on the forehead. I should have had a V8!

Of course the other part was it was a work in progress and sometimes I couldn't risk keeping it floored if the AFRs would creep past 12.5 and into the 13s (scary lean territory in boost).

So yeah I think the secondary tune is good now. I'd like to try some custom drilled .20mm sec slows though. I just need to get those awesome quality drill bits and find a supplier of blanks in the B42/55 style. Or real Hitachi or Nikki blanks would be better. Anyone know where we could get any?

LizardFC

iTrader: (1)

I'm still checking in on your thread even though I haven't been able to play with mine for awhile. My car is currently off in another state getting a cage made. Safety first, right? The last time I had it together and running, I was still having issues with oil inside the turbo. Once I get it back and put everything back together, I'll try and sort that out and start fooling with the carb again. Thanks for keeping such detailed accounts of your findings. It's a lot to read through but it's helpful stuff.

Your findings with the filled slow air bleeds sound like the same results I had - Inconsistent AFR's and sporadic idle issues after boosting. I'm glad you got yours working without plugging them.

One of the last things I fooled with before I started stripping my car for the cage was modifying the secondary throttle linkage. I made it so the secondary plates opened really early. I think they began to crack open at less than 25% throttle, which seemed to keep the transition from light to full throttle a lot smoother. Probably terrible on gas mileage, but my car is (almost) a race car so I don't care. Might be something for you to toy with if you haven't yet.

Keep up the good work

Jeff20B

iTrader: (13)

I'm glad you're still checking up on my progress. You'll be way ahead of everyone else when you get your car back.

The secondaries are set to open at about 60% throttle normally on FB carbs (SAs open at more like 70% making them less desirable). I'm not sure I'd need to adjust the 60% if I remember to floor it more often. Now that I can safely, that is.

Jeff20B

iTrader: (13)

Update time!

So I got the high quality Japanese micro drillbits and drilled a set of brass blanks for the secondary slow air bleeds. The smallest I've ever done: .25mm through solid brass!

Then I took a set of 60 main primary air bleeds and drilled them out to 70. Actually 67 to 68 which is exactly the same size Nikki uses on their 70s!

Then I installed a set of Hitachi pilot jets which were originally 40 but drilled them out to 80 for the modified main secondary air bleeds.

This combo is untested as it was getting late and starting to rain, but I think it will work seeing as all the sizes of these freshly drilled parts are very similar to what was already in there and performed great.

Jeff20B

iTrader: (13)

And they did perform great! It seems the tune it perfect now. The carb did exactly what I wanted it to do, from breaking them loose on dry road with just a flex of the ankle, to returning back to idle without a hitch. Or rather without a glitch. No secondary delay either, which has plagued me literally from day one. It's gone now. I floored it at 3k and that was all it took to put a smile on my face. Perfect.

Final tune 25.6mm venturi carb
pri short slow: 105
pri long slow: 42
pri main air: 67
pri fuel jet: 124 (summer tune is 13.0 AFR at part throttle)
sec main air: 80
sec fuel jet: 140
sec long slow: solder filled
sec short slow: 25

Qingdao

iTrader: (3)

Do what I did. Cut 12a sized intake holes in a plate of steel (or Al*) one side and 13b sized intake holes in the other side. Drill and tap 12a bolt holes in the 12a side of the adaptor; and drill strait through holes for the 13b.

If you use 1/4" plate steel an S4 turbo will clear (with room for a heat shield) after adding 1 1/4" of space from the turbo to the engine. You MIGHT be able to clear the turbo if you use the Mazdatrix custom intake plate.

The only other problem is going from 4 to 6 ports, but it'll work. I've been running my Edelbrock for a year on a 6 port mani to a 4 port engine. I bet I wasn't getting 100% of the power I should have been getting, but it worked.


*Al precut manifold
https://www.mazdatrix.com/getprice.a...tnum=MZINT-13B

Jeff20B

iTrader: (13)

Since this is where I've posted most of my hogged out boost prepped Nikki pictures, I figured I'd post this here. It's the latest carb I did a complete work up on. Ugly on the outside, pristine on the inside. From a very low mileage car probably driven in the salt belt.

So much power it breaks them loose in primaries. Drives really well too.

I don't think posting pics to the forum works very well anymore because it always gives me the rel=lightbox thing, so feel free to click the thumbnail.

Note: the text I added is for those needing a visual reference for where the air bleeds and jets go. Notice the nickel plated short slows get moved over to the primary side and drilled to 118.

Sorry about the messy work mate surface. That's just how it goes sometimes.

Qingdao

iTrader: (3)

Ok. I can chime in now.

I have an S4 turbo, S5 street port block.

I've got ~24.6mm primary venturis

Pri. short : 118
Pri. long : 46*
Pri. air : 70
Pri. fuel : 118
Sec. short: Solder filled
Sec. long : Solder filled
Sec. air : 140
Sec. fuel : stock


I'm golden under boost, but when cruising in the primaries I am really lean. Like 16-19, which is unnerving.

I'm thinking I should go ahead and move up to the next larger bit in my collection on the primary short. I have a 1.27mm bit. Does that seem like a correct move?

Jeff20B

iTrader: (13)

127 is too big. I would try a 121 or 122 first.

You can't expect to fine tune a carb with such wide gaps between drill bit sizes.

Ah heck 127 is all you have. Go ahead and try it. But know that you are probably going to ruin that set of jets by drilling them that large.

For example the 25.6mm carb listed above was recently updated just on the secondary side but needs to be test ran. Anyway its primaries are still at 124. Think about that. 124 works with venturis at 25.6mm. Your venturis are a lot smaller. Do you really think 127 is a good idea? I don't. I think 121 is a good place to start but you don't have that size.

The carb pictured above got hogged out to 24.6mm like yours and it did fine with 118 drilled jets so something is up with your carb.

Another option is to get a couple sets of jets from jetsrus. Or a couple of drill bits from drill bits unlimited. Create Your Custom Drill set:-Sizes: .038" to .0785" - Drill Bits Unlimited.com

Oh and your sec shorts need to be drilled out to .35mm or .37mm. But don't expect the drill bits from drill bits unlimited to handle it. You will need to get the cheap chinese drill bits like I got earlier this year. They can handle solder.

wankel=awesome

Small jet increments make big differences in the nikki primary circuit. 121 might be a tad smallish but it should be close

Qingdao

iTrader: (3)

Yeah, 127 didn't work... I'm still kinda under the impression there is something clogging some circuit somewhere. Cause the leaness got better after I cleaned it out some the other day.

Any ideas?



I haven't messed around in the secondaries at all. Untill a couple of minutes ago. GOD DAMN. Like a hole nuther car is bottled up in there. Regardless I'll check my bits to see if I have a .35

Jeff20B

iTrader: (13)

Maybe some rust chips got down into the idle circuit. The idle circuit is just a calibrated slot milled into the cast iron baseplate. When the factory is done they press in a sold brass plug with some kind of epoxy. You have virtually zero access to it other than through the small drilled hole that connects it up into the main body. I used a sewing pin with a tiny hook I bent into the end of it to clean out Jingkun's carb. It got all the rust and glass beads (or I guess it was media plastic beads) out of both slots. It took a lot of work and lots of compressed air but all the crud eventually came out. His carb actualyl idled after that.

Your carb might have a problem with something clogging the transition circuit as well. That is handled by the long slow 46 air bleeds. Make sure the tiny .46 hole at the bootm can be seen through. Don't poke anything in them unless it is like a tiny bristle from a small brass wire brush or something. This calibrated hole is the most sensitive part of a Nikki. I know this becaue I did extensive testing earlier this year. The Nikki is so sensitive to these things that you can tell a difference between two sets of factory 46 long slows based on how they make the carb run. The transition circuit actually overlaps the idle circuit and the main primary circuit and any changes to these long slows will affect all three circuits noticeably. The carb's personality changes. Sometimes for the better and sometimes for the worse. You can also see it in the wideband. It's literally based on how much wear the factory's micro drill bit had when they were drilled. Like less than a hundredth of a mm makes significant changes. So you really need to be sure yours are clear.

The .35 is about the width of a human hair. Well, maybe a little thicker. It's very small.

Qingdao

iTrader: (3)

Ok, we're getting better. I didn't even think about cleaning the base plate. I didn't even think there was anything important in there.

So I took a single strand of copper wire and cleaned out that near impossible to get too hole. Then I shoved a latex glove in the bottom of the base plate and blew compressed air in it. Out came a sputter of some kind of varnish.

The carb idles at the fuel rate I want it too now.... turning the mix screw actually changes things now.

I think I still need to run a little more carb cleaner through the tank to take care of the finner parts in that orifice. but 11.5 to one is much better at idle. The engine just sounds so much happier.


Unfortunately I don't have a .35mm bit. The smallest I have is .51mm, but at least I have the vise so all I need is the bit. I haven't installed my 118 nickel plated primaries yet, but the 127's don't seem to be hurting anything at the moment.


EDIT: Jeff where do you get your 13B manifolds?

Jeff20B

iTrader: (13)

The .51 might work out for you. Remember I was fighting a secondary delay which you probably don't have to worry about.

As for your primaries, 127 is also probably ok for you. As long as the mixture screw does something as opposed to nothing and puts the idle circuit to around 12.2 (where my engine likes to idle best - yours may vary) and allows some adjustment up and down, you're fine. I have good results with primaries drill as small as 105 on some carbs, but that is only on the ones I cut the venturis really big. Perhaps too big and I needed to compensate some how. And because the long slows are off limits, it had to be the short slows.

The NO hitachi manifolds are probably the best, and they are rare, being from Japan only. The slightly more common REPU manifolds are second best but they require more work to fit a Nikki. The 76 reverse runner manifolds don't seem to work with an S5 turbo due to runner positions.

None of these will fit your FC engine.

You're better off finding a 79-80 12A mani and adapting it or better yet an old school RX-3 or RX-2 and adapting it.

erick31876

iTrader: (8)

I might have a extra nikki manifold laying aroundfor you Qingdao

Qingdao

iTrader: (3)

81-85? I've got an 84' and its just sooo small. I need my SA's manifold for my SA; so I'm not gonna rob my 79' for a good intake. Just wondering what my options where.


Maybe with the smaller intake holes I've got more low end?? I guess thinking that will make me sleep better.

Cookboy

iTrader: (10)

I just saw a nice looking early 12a mani on eBay. It's in the UK. Probably just over $100 shipped.

erick31876

iTrader: (8)

I think 79-80, and a couple later ones, I have a few of them, pm me your info, and I can send you pictures