On ‎25‎/‎04‎/‎2017 at 08:44, dvderlm said:

Gauge thinks my engine will do 25999 rpm!

Maybe connection to coil negative is affecting ignition advance too? 

It's not affecting advance , but one is not meant to use coil negative or a crank sensor for speed input to the AFR logging gauge.

http://aemelectronics.com/?q=products/wideband-uego-air-fuel-controllers/wideband-failsafe-gauge

Specs tab.

 

  I have not damaged the gauge but the readings get skewed.

 

...

 

I have now got a Hall sender that can detect a magnet passing. If I glue this magnet to cam pulley or crank pulley or flywheel and fit a bracket.

Will magnet stay on, will bracket foul the aux belt?

Why not try piggy-backing the ignition Hall sender in dizzy?

If it's good enough to set ignition point it's good enough for logging engine speed.

Maybe via a diode or opto-isolator...

 

Edited 13 May, 20179 yr by dvderlm

Engine won't start if gauge speed input connected to the  dizzy's Hall sender.

 

Not sure about the glue to use. Magnets must be correct polarity too. The neodymium magnets on my oil filter have stayed on for 4,500 miles, but they are just resting not spinning at 7000rpm.

If use cam pulley that's at half of engine speed, so 3500rpm max, would need two magnets and tell gauge one pulse per revolution.

Maybe two rubber bungs in the cam pulley holes for balance. No, though there are 6 holes, they aren't round.

 

cam pulley

 

More fun. Engine died on way home, loss of sparks.

 

Thanks to the 3 wonderful people who stopped to help, driving back around a roundabout after seeing me, pushing car off road. One who even confirmed no spark at an unscrewed plug for me.

 

Had to switch back to Ye Olde Dignition Box.

 

I had disturbed the cable to Hall sender when i tried piggybacking.  This meant a very thin wire flexed and snapped inside the insulation. Maybe it's been flexing for a while, but speed bumps and Cambridge roads did for it.

Think I'll apply some Sugru to that joint...

 

But anyway - using the Dignition reveals just how much timing advance matters. It drove, but felt underpowered and the AFR gauge recorded a richer biased log around either side of 14.7, while with the Amethyst it feels good and is more lean biased from 14.7 towards 16-17.

 

Both have about the same proportion of 18:1 - I think - and the removed plug looked light tan not scary lean white. I had switched to some 3 pronged Bolk plugs to check this anyway. 

Flaxcar 43013z.

http://webcatalogue.klaxcar.com/catalogue/en/product/43013z -link dead. These.

http://www.mister-auto.co.uk/en/spark-plugs/bolk-bol-c051150_g686_a999BOL-C051150.html

 

Edited 15 May, 20179 yr by dvderlm

Well Dave Andrews, JC Nieutin and Des Hammill reckon 40F8 in a Weber for 1300 and a clan imp engine driver says small idle with good advance works for 875cc and 1040cc with overlap cam.

F8 has a single 1.20mm emulsion side hole.

40F9 is richer still due to smaller 1.00mm side hole.

But there must be a top air corrector for the idle, fixed size in Weber.

 

DHLA has fixed size idle air correction for the emissions carb models at 2.2mm and no upper holes. But none-emissions carbs have upper holes in the 7850.x tubes, usually 4x 1.30mm or 4x 1.25mm.

The richest 7850.8 has4x1.40 uppers and  4x0.5mm lower holes.

So, I'm going to make idle emulsions richer than 7850.2 but not as rich as 7850.8 by using .5s and making a single small lower hole.

As Dave Vizard explains, its emulsion holes nearest the fuel jet that affect top end of rev range

That should still  idle nicely at 800rpm, but continue to provide fuel upto 3000rpm with the 30mm venturis and 6 transition holes.

 

Jb weld and needle file and micrometer

Edited 26 June, 20179 yr by dvderlm

Changing to these means removing the inlet plenum. It does not cover the shared idle holes, but makes the main circuit a little over rich. Used a dash cam to match AFR log to joining  motorway acceleration lane.

 

This is 10.6 seconds Wot acceleration with car packed with camping stuff 23 to 70mph with one gear change. 12:1 with 119 main.

 

Compare this to trumpets (but with 110 main)

 

 

So needs smaller main or bigger air corrector or larger chokes. Maybe 115 main with plenum?

 

Still 18:1 with the custom idle holders, but I can greater idle vacuum at about 900rpm.

Much belter mid-range revs and 25 to 40mph too.

 

And tried 200 air correctors on main jet with trumpets, which sound glorious.

 

7.9 seconds 10 to 60+ something mph with one gear change...  really need to sort out an accurate the RPM sensor. I suspect the peaks are harmonics of the inlet tract resonance.

 

 

On 09/07/2017 at 16:29, dvderlm said:

really need to sort out an accurate the RPM sensor.

 

 

 

Why did I just get a mental picture of you driving down the road with a CRT oscilloscope on the dashboard, hooked into the hall sender?

Too bright for CRT in the daylight. Too noisy for engine experiments at night.

 

But an LCD oscilloscope could work...

https://www.amazon.co.uk/dp/B01LWK49W3

 

Shame a 2.4inch TFT display won't fit in a 52mm gauge pod. Would look well De Lorean flux capacitor.

What I'm trying to grasp is why it goes richer for a small fraction when the throttle plates are closed after full chat.

I think the fuel has more momentum than the air so carries on into the combustion chambers because of cam overlap sucking but the airflow is cut off more easily so fuel starves of oxygen.

 

My lambda sensor is about 70cm from the exhaust valves but manifold pressure sensor is 120cm from 4 inlet runners.  i suppose the inlet is nearly closed system between throttle plates and inlet valves, though

all connected via thin balance pipes and accumulator to smooth pulses for the ignition. But it looks like the lambda takes about 0.03 seconds to see the impact of throttle plate position change.

Is that just over 2 engine revolutions at 3000rpm or 8 to 10 cylinder inlet strokes?

Edited 11 July, 20179 yr by dvderlm

Custom idle holders have single 0.96mm lower hole. I was going to increase to 1.2 to 1.3 mm if too rich. It is touching 7850.8 for richness - it has fewer but larger upper holes.

I suppose it can supply quite a glug as the throttle plate heads to closed position passing the progression holes, so going back to 36 idle jet in this holder.

 

Another table of AFR ratios from Pumaracing with useful CO equivalence for petrol combustion.

Definitely too rich at 12:1 with 119 main and 200 air in 30 chokes. 

Think I might start drilling out the 110 mains by 1mm at a time... 111 is 1.8% larger, 112 is 3% larger,  ... 114 is 7% larger and 119 is 15% larger than 110.

 Except  this clashes with Hartman, so might need a g-tech pro or seat of the pants, stopwatch and dashcam.

CO %	A/F Ratio	Lambda	Notes
0.00	15+	>1.00	Cruise mixture for good economy
0.25	14.7	1.00	Stoichiometric mixture
0.50	14.5	0.99	Standard FI engine idle reading
1.00	14.25	0.97
1.50	14.05	0.96	Standard carb engine idle reading
2.00	13.80	0.94
2.50	13.50	0.92	Race engine idle reading
3.00	13.30	0.90	Standard engine at full throttle
3.50	13.10	0.89	Standard engine at full throttle
4.00	12.90	0.88	Good power with reasonable economy
4.50	12.75	0.87
5.00	12.60	0.86	Race engine at full throttle
6.00	12.25	0.83	Too rich

Edited 17 July, 20179 yr by dvderlm

So the gear change makes interpreting harder because of richness when throttle plates closed. This is 5 seconds from around 20 to 50mph with drilled 111 main and 200 air .. that went from 13:1 with 110 main  to 12.3.   

 

So bigger air corrector ? Bigger choke? Smaller main and less holey emulsion tube than  economical .10?

 

Think it shows the pump jet squirt needs lengthening too.

I had bigger air correctors and thinner richer tubes  in the 34DMTR with 25/26 venturis https://www.clubpolo.co.uk/topic/332122-weber-34dmtr-2nd-stage-stumble/

That felt good before piston ring change. Twin choke progressive though. Wonder what the mid point of those two is?

 

126/225 ?

Lowered float 1mm.

Needs higher engine load.. 40 to 60mph in 3rd gear in 3.9 seconds ... same jets 111 main 200 air

 

A near perfect 12.76 AFR.

 

And the 12.3:1 occurs at 5000rpm and above.... so yes bigger air correctors or less holes near main jet in the emulsion tubes.

( slightly cooler air temp?)

 

That's around Subaru Impreza 2.5 acceleration...

http://car-videos.net/performance/speed.asp?Speed1=40&Speed2=60

 

 

 

Edited 25 July, 20179 yr by dvderlm

Alpine A110 use twin Weber DCOE on high compression 1300cc (or 1600). 

This guide reckons trumpets need 0.05 bigger main (and then 0.10 bigger air corrector if too rich after main jet increase).

 

Also has some DCOE 45 jet suggestions for 34mm or bigger chokes. A110 1300G has 33mm chokes as standard.

 

 

 

Maybe change from 12.7 to 12.3 above 5000rpm  is okay... this is from the Megamanual http://www.megamanual.com/begintuning.htm

megasquirt guide. Top row is Wide Open Throttle.

 

So moved to 38 idle in my custom idle holders now with two lower holes approx 0.95 and 0.35 mm diameter-about same as a single 1.0mm hole? That feels good. Still 18:1 at 900rpm but will pull to 3000rpm.

For a laugh tried 55 in these holders which idles at 12:1 and drives at 10:1. Opened all four idle bypass screws a tad (1/16th of a turn from seated) and it will idle at 13:1 at 1200rpm but will then go as low as 9.6:1

in motion, which  sounds like a dragster, but is actually sluggish.

 

So I know the limit of what to fit for turbocharging if I get that far.

http://speedtalk.com/forum/viewtopic.php?f=1&t=36738&start=30#p561842

 

Harry explains that you need smaller idle with more transition/progression holes.

 

 

Used a Tesla Electronics g-tech pro accelerometer 

http://gtechprosupport.com/support/manuals/ProMetric.pdf

and guessing that the car weighs 820kg = 1820 pounds with me and fuel and stuff in it - I got a horsepower reading of 94 DIN twice with 30mm chokes 111 main 210 air in .10 emulsion using trumpets at around 13:1 AF

It's supposedly accurate to +- 3hp but i did not do average of several runs and will be lower than dyno due to aerodynamic drag.

Need to compare with using 200 air and with plenum instead of trumpets.

 

 

On ‎24‎/‎08‎/‎2017 at 23:36, dvderlm said:

will be lower than dyno due to aerodynamic drag.

Ah, but it measures (rear lol)wheel horsepower not flywheel horsepower,  according to the documentation. "includes loss of power through the drivetrain which is usually 10-15%". So is it adding 10 to 15% to show 'H 94' or is actual flywheel horsepower 105 to 108 ish?

 

 

 

 

 

 

Well a wideband AFR is good, but I need a 4 gas analyser. Failed MOT on emissions. An amazing 0.135% CO with my custom idle holder and 38 idles, but excessive unburnt HC, so that's the 18:1 AFR at idle reading explained. It was 1.25% CO and ~300ppm HC last year.

 

The CO limit is crazy 4.5%, HC limit is generous 1200ppm. I got 1553 ppm even with 2000rpm recheck.

 

Which has to be ignition fault or timing off because the idle jet is only slightly up (36 to 38) and the idle holder is leaner than the .7 tube that passed last year.

Can't be that far off because progression and cruise is fine. 

 

Today's investigation found number 3 spark plug lead not making good contact, and trumpet on number 1 cylinder lost a lower fastening screw, putting all the balance out. So, did a complete reset. Experiments with 36 idle just got racing or too hot, so more timing advance needed. So now switched to using 48 idles in .1 holders, 4 vacuum guages (1 per cylinder), a Gunson colortune, the AFR guage and a sonic screwdriver I set the engine to lowest stable idle I could manage. Then played with the Amethyst ignition timing at 500, then 1000, then 1500 rpm with no vac advance to find highest vacuum on all cylinders and wound back about 1inHG. Number 4 cylinder feeds the brake servo, so it is always less. Then added 9 degrees of vacuum advanc allowing me to reset mixture screws tighter to reach 14.3:1 AFR, then adjusted idle bleeds to get average 14.6:1AFR and rechecked balance. It sounds crisp. 900 to 2000rpm is smooth (no load) but touches 12:1 which might be pump jets.

 

 

Edited 28 October, 20178 yr by dvderlm

Balance cylinder #1 and #3 to about 1.7cfm at idle.

Picked after passing MOT. Emissions now 1.987% (=13.9:1 AFR) and 1078ppm of HC.

So I can go leaner than 48 idle in 7850.1 holder. Below 2000rpm and light throttle it will manage 14.7:1 or leaner but above that the crossover from idle jet to main circuit means W.O.T is a little richer than I'd like but not awful.

 

Should be more efficient. I'll see what cruise is like on a motorway.

 

 

 

 

Not been on a motorway but steady 60mph on A-roads is sometimes pushing 12.5, so have gone down to 45 in 7850.1 holder and unscrewed idle just over 1/4 turn. The idle speed's a little lower and speed screw adjustment increase stalled it,  -  especially with all electrics on in the cold - but gently pushing accel pedal to drive off doesn't, maybe it's operating the pump jet. Wish I could adjust the Amethyst by 0.5 degree at 500rpm. Had to up the vacuum boost to 10 degrees at 15inHg (-7.5psi).

 

Idles at 14.8:1 AFR, drives at 14.7 to 15.4 on light throttle, 12.6 to 13.2 on hard acceleration but will get 18:1 on overrun.

Might need another session with vacuum guage and colortune, and twist the dizzy to get that 0.5 degree adjustment (0.25mm at circumference!)

 

 

 

18:1 AFR on overrun has gone since I fixed cylinder #2's misfire after icy weather!

So the relatively large vacuum with closed throttle (at overrun) must have been acting on the idle circuit which was partially blocked.

 

The idle became richer of course - I simply retuned for same exhaust gas temp as the other cylinders. Need to rebalance.

Maybe smaller idle still?

 

 

 

Have only very quickly scanned over this thread, and think you may have a load of replies from someone else missing too...

 

A friend fitted a wideband AFR gauge on his BMW, which was on stock injection and running healthy, and half the time that idled leaner than 18:1... at the end of the day theres guideline figures to go by, but also if your engine runs nicely and pulls smoothly etc its not something to get drawn into hugely IMO, if it reads lean, but idles nice and smoothly or whatever just go by what the engine likes. 

 

Obviously under load/wide open throttle you need to take more care so it doesn't melt itself... but dont get too hung up on figures in low load/idle situations.

That's very interesting. I suppose injection can turn off a cylinder every so often and it will still keep turning thanks to flywheel.

Low volumetric efficiency at idle, so ignition advance matters to get max piston force by 14 degrees ATDC. 

 

Thread summary:

There was an MOT fail issue I caused in trying to beat the lean AFR (it had passed with smaller 26 chokes last year before I had the gauge with 36 in .7 holder). This was from using a lean-but-richer-a-high-revs custom idle holder and the small 38 idle jet with not very advanced ignition timing (still 9-ish degrees Btdc) was creating a misfire with unburnt HC. No pops, no shudder, no visible symptoms. To get through the test  I went up to 48 idle and a fairly lean .1 holder and throttle plate closed tight, but crucially increased idle advance to about 14/15 degrees Btdc. Do the same with the rich holder and idle speed was too high for my liking - but probably legal for emissions). There was also another problem that cylinder 2 idle circuit had a partial blockage, perhaps why someone sold the carbs, guessing it probably had killed a Lotus Esprit Turbo.

 

I'm now using the same fairly lean holder (7850.1) with a 43 idle jet as near the midpoint of 40 (surging lean at 2000rpm) and 45 (too rich at 2500rpm), because this range is not really adjustable with a screw. Well the idle air bleeds could adjust, but those are best used for cross cylinder balancing (no more than 2 bleeds open at all) so there's maximum suction on the idle circuits.

 

It's far smoother when setting off from rest, appears to cruise at ~15:1,  70mph motorway at 13.5 to 14.5:1, but W.O.T (once pump shot exhausted) has climbed.  So back down on the air corrector again from 210 to 200 or less to shift the point the mains come into action and sustain the fuel flow over 5000rpm.

 

All carbs are compromise.

Edited 15 January, 20188 yr by dvderlm