Needle Modification for L20b SU's

[mtngoat]

Hi Folks

well I finally decided to get around to modifying some needles so my L20b actually gets the fuel it needs when it needs it, instead of being an unhappy mix of out of tune mixture.

 

I'm running the M66 needles and it exhibits the classic signs of using these with a 2L, in short, if you have it rich enough for the freeway it's way over rich everywhere else, including idle. I have a WO2 gauge so I know exactly what I'm seeing. Right now it's set up way rich at idle and part throttle so at freeway cruise I'm running 13.5-14.5. At idle it's pegged at 10, and of course I had to dial the idle speed up just to get it idle halfway decently.

 

So I went and ordered another set of M66's and the DQ article on modding needles. The theory is simple enough, you sand down the upper cruise and WOT stations starting at 7 to richen them up, and then when you install, you can lean the entire range down to where it needs to be. With the richer tips, the profile is no longer way rich at low throttle, to lean at the ends, and that allows setting them up properly. Makes sense, so far so good.

 

Then a couple nights ago I sat down with the article, the new needles, a dial caliper, and Excel. First I find that when I measure my needles, they do not match the profiles listed anywhere for M66 needles. I went out and swapped the new ones into the car, and then measured the ones I had been running. Same result...identical to my first set, doesn't match the supposed M66 numbers from Nestor Moya's measurements, or anyone elses.

 

Confused by that, I decide OK I'll have a look at the numbers for the mod shown in the article, and ignore, for now, that my needle profile doesn't match the 'stock' values listed in the article (which match Moya's). I enter the 'stock' numbers, I enter the changes listed (-0.001 at station 7, -0.00x at station 8, and so forth), and I enter the profile the author measured for his needle after the changes listed.

 

What do I find? I find that when I subtract his listed changes from the numbers for the stock needle, it does not match his measurements for the modded needle. It's *way* off with a tip thickness of 0.033 or something, not the 0.061 or so listed as his result.

 

To confuse matters even more, when I average the measurements I took from my four needles in order to arrive at a reasonable approximation of my profiles given measurement error, I find that they are *already* nearly identical to his 'modified' needle! And, the new needles behave exactly as the old ones did on the road.

 

Needless to say I now have a quandry on my hands. What I see driving my car as it is now matches all the characteristics noted for a set of M66's, but my unmodified ones don't have that profile that I can tell. Worse, my needles are already darned close to the end result of the mods listed in the article.

 

I understand the theory and it makes sense, the mods are simple enough to do, and I have a spare set of needles. The ones I do have definitely are too lean at the tips, same as the problem in the article. So I'm inclined to proceed and mod the needles as I intended, beginning at station 6 or 7, and thin them out a bit by some amount yet to be determined.

 

Thoughts?

[KlassicMotion]

I'm not sure how much this will help;

 

I haven't dived into the wonderful world of Hitachi SU needles yet, but I dipped my toes in the water with my British cars....  Can you use the British needles in the Hitachi SU's?  If so, there are lots of needles on the market for those.  I can't remember off hand if the Hitachi was similar to the HS2 or HS4....

 

Here are some of the resources I've bookmarked:

 

I love these two charts:

 

http://www.rrocncr.net/technical/SUChart/SUChart.html

http://www.mintylamb.co.uk/suneedle/

 

http://www.7ent.com/pages/articles-tech-tips/chart-carburetor-needle.html

[mtngoat]

Thanks Kyle. Yup I looked into that before pursuing the modification route, there really aren't any great needles for this combo, which is strange given how common it is.

[KlassicMotion]

Hmm.... I wonder about the needles used on Volvos....

[Eagle_Adam]

have you looked up Ztherapy? 

[datzenmike]

Hmm.... I wonder about the needles used on Volvos....

.

Volvos are British SUs.

[mtngoat]

have you looked up Ztherapy? 

Yup, that's where I got both pairs of M66's. To my knowledge there is no source for needles custom designed for the L20b/SU setup. Which is surprising, given that this was one of the most common swaps done for a long time.

[KlassicMotion]

.

Volvos are British SUs.

 

I wondered, but I hadn't Googled it yet!

[Sealik]

Yup, that's where I got both pairs of M66's. To my knowledge there is no source for needles custom designed for the L20b/SU setup. Which is surprising, given that this was one of the most common swaps done for a long time.

 

Custom profiling of needles is somewhat needed for any engine ...being that no two are alike...<... modifications etc.

ZTherapy suggested RA needles for my LZ23/46mm SUs....not even close looking at my A/F gauge.

 

That said....I did manage to get 'cruise' to 14-15 and WOT at 12.5

Buff needles, measure, install, observe ratios, remove, buff...install...etc....many times.

Initially I wasn't sure how much to remove to affect A/f ratios

 

Was a compromise though....being that I couldn't add/fatten the needle to lean it out at idle.

 

Idle is still rich at 10.5....but...I'm not sitting in bumper to bumper traffic. Choke it, flash it up......slight warm up....gone

Plugs look good unless I'm sitting there for 10 minutes doing 'diagnostics' on the SUs and or.

 

 

[mtngoat]

There's some good ideas there...I can only get my drill press down to 300 RPM and I like the hand drill idea.

 

It looks like you have a template there for comparison to a profile target? I  like that idea.

 

I get what you're saying about adding material to lean it out at idle, but if you think about the point of the article it's to richen the rest of the needle to bring it closer to the rich setting at idle, then lean the entire range by moving the seat up when you're dialing it in. He claimed to gain an entire turn leaner.

[mtngoat]

Resolved the math error in the article. the modifications listed omit a 0 in a decimal place following the decimal point for stations 10-14. What is listed as -0.013 should be -0.0013. 

This still doesn't explain why my M66's seem to match his modified M66's, though.

[Sealik]

There's some good ideas there...I can only get my drill press down to 300 RPM and I like the hand drill idea.

 

It looks like you have a template there for comparison to a profile target? I  like that idea.

 

I get what you're saying about adding material to lean it out at idle, but if you think about the point of the article it's to richen the rest of the needle to bring it closer to the rich setting at idle, then lean the entire range by moving the seat up when you're dialing it in. He claimed to gain an entire turn leaner.

 

Ha ha...had to use a 'fresh' broccoli rubber band to get the correct RPMs.... :D

 

No template.. target unknown...just a record of the amount removed from the needle as I progressed.

Wanted a somewhat smooth transition through the stages being that I had to remove a considerable amount in the middle of the needle, to richen it up.

Raising the jet at idle just moves the stages to a different RPM/throttle position....etc.

 

I found the fattest needle I had available and started 'whittling' away to get the best possible ratios....not perfect but good enough for the first attempt

[Just Joel]

Anyone have a pdf of the dq article mentioned?

[mtngoat]

well, i went and did it. 30 bucks...down the drain, or success? 

 

I started a station earlier than the example given my guesses of how the gauge reads as i put my foot into it, so the mods begin at station six. I also went more conservative than his example, I used a step size of .0003 for a total of -0.0027 at the last station. 

 

I started at 6 and just hit it lightly for a few seconds, figuring that was enough for .003..way below the range of my calipers!

 

then twice that long at 7, three times as long at 8, then I hit all the rest with the same amount of time. after two subsequent passes like this beginning at 8, 8 measured out only slightly above what i wanted, and the rest were short. I then worked 9-14 figuring some would come off at 8 anyway from doing 9. one more pass had 8 dead on and 9 close, so I moved to concentrate on ten and higher. 

 

I repeated this a few times and it seemed to work. the last stations took about 4-6 more measuring and sanding cycles. feeling confident now, the second needle went faster and I think they're pretty close to each other. I'm going to measure both of them twice later on, and see what I get. If I need to touch em up i'll of course have to richen the leaner one!

 

then, I'll work up my courage and put em in, likely tomorrow night. I've spent so long dreading taking sandpaper to these things it's anticlimactic now...all that remains is actually installing, tuning, and trying em out!

[mtngoat]

Anyone have a pdf of the dq article mentioned?

no, i bought the paper version. I saw your earlier thread on this from googling the other night!

[Just Joel]

Yeah, I profiled a set from scratch with no particular target in mind other than what the gauge was telling me. It was a slow painful process of trial and error but I'm getting close to a good set. I didn't know (or perhaps forgot) that dq did an article on this so I'd like to compare to the info they came up with. What issue is it? I know that all the articles are pdf'ed somewhere now.

[Just Joel]

In addition, I have compiled a collection of stock datsun 38mm needles and have found the m66 to be far too lean (compared to others like the m77). I have also found that the aftermarket needles provided by z therapy are no where near the original m66 profile. They are however a nice "fat" needle to start profiling from ;)

[mtngoat]

volume 4, issue 1

 

yes, M66 is crazy lean. with a nice idle at 12 or so I'd be pulling 16+ on the interstate. Not good! Having burned up one L20b already by just going with stock needles and setting it to a nice idle by ear, I didn't feel like going that route again with a fresh engine, so I picked up a WO2. Expensive, but the most useful tool I've ever seen for carburated engines. it's amazing. never going back. 

[KlassicMotion]

I picked up a WO2. Expensive, but the most useful tool I've ever seen for carburated engines. it's amazing. never going back. 

 

What is this you speak of?

[mtngoat]

The Wideband Oxygen Sensor, i had no idea such things existed until a few months prior to buying it. Love it. 

[KlassicMotion]

Ah, Wideband... got it.  Thanks.

[Just Joel]

Yep... I tune with an Innovate WO2. I've found that these motors are extremely tolerant of lean condition but it's deafanantly not optimal for performance ;) I've see numbers in the 16 range all the time at cruise with no hesitation or detonation/pinging but you can feel the performance difference.

[KlassicMotion]

I've only heard of Wideband, never used it.  Is there a good place to find a crash course in it?

[mtngoat]

a zillion online. it's a fancy oxygen sensor which tells you rich or lean. it lives in the exhaust pipe. it's fancy in the more modern sense because while old ones were nearly binary devices...rich, lean, and not much else...the modern ones are very precise.

 

you get the sensor and a gauge, generally as a kit. mine reads in tenths from 10 to 17 I think it is. I have numerical readout plus a full sweep three color LED ring around the exterior which acts like an analog guage. 

 

mine can be set all kinds of ways but I use ratio. 14.7(:1) Air/fuel is a perfect chemical mixture in the frictionless ice world of perfection. It may not be best power, or economy or whatever, but in chemistry it's when they're in perfect ratio. 

 

So I paid about 160 for mine. it comes with the bung, which you have to weld into a hole in the exhaust system. mine is not to be closer than 24" inches I think from the ports. then you connect up to the gauge, and you can tell what the heck is going on any time the engine is on. rich, lean at idle? rich all the time? rich at idle, lean at cruise, too lean when you stomp on it? all in real time. adjustment of any screw or jet is immediately visible, assuming you're in range of the readout, of course. 

[mtngoat]

at some point i really need to get around to installing the pyrometers I picked up. I'm going to do cylinders 1 and 4. this will finally, for me, settle the question of how lean is too lean by resolving what the exhaust gas temp is, correlated to mixture.