Finally fitted a wideband AFR lambda gauge to the bung I had fitted on the rolling road day.

 

Before I had  the bung fitted I had used lambda temporarily attached at tail pipe with some flexi and assumed the super lean idle reading was air dilution though being like 6 inches from the outlet.

Can't detect any exhaust leak.

 

But at a point maybe 20 inches from exhaust valves it reads 18:1 at warm idle at ~900rpm.

(pull the choke and it 'll hit 13:1 so it works).

Go for a drive and it'll go as low as 10.5 at initial hard acceleration (think I need to reduce pump volume or shorten duration) but cruise at 14.5 to 16.

 

Anyone else got unbelievable AFR?

Is it just flowing too little (4cfm per cylinder) to read?

 

Twin dellorto 40

with 48 in 7850.1  or 51 or 55 in 7850.5 idle holder.

 

 

Edited 17 April, 20179 yr by dvderlm

It looks like it's too lean at idle. It depends on the engine. The ideal for petrol is 14.7. Yes lean up the accelerator pump volume and increase duration to give the best acceleration. Cruising at those levels when the engine is less loaded looks good to me. It's either put it on the dyno with some changes or just get a feel for it. Pull the plugs out at the various times and look at the colour. They will tell you if you are lean or rich. 

According to http://www.onedirt.com/news/how-to-fine-tune-your-carb-using-holleys-tuning-dvd-and-wideband/

large overlap camshafts can give false reading at idle.

 

The following are typical A/F ratio ranges for various engines that should result in good performance. These are suggestions based on Holley’s experience. Your engine may require a leaner or richer mixture than those suggested below.

Stock and Moderate Performance Engines

Idle and Cruise: 14:1 to 15.5:1

Moderate Load: 13:1 to 14:1

Wide-Open Throttle: 12.2:1 to13:1

Large Cams and Race Engines

Idle and Cruise: 13:1 to 14:1  (Large overlap camshafts can give a false lean reading at idle)

Moderate Load: 12.5:1 to 13.5:1

Wide-Open Throttle: Typically 12.5:1 to 12.7:1 at peak torque and 13 to 13.2:1 at peak HP

Supercharged/Turbo-charged/Nitrous Applications

Boost Conditions: 10.8:1 to 11.8:1 for safe operatio

If the Gk cam (052101109h or b ?) is not a regrind then it has 3 degrees overlap.

Hardly 'large' so maybe vacuum leak?

Perhaps throttle spindle or choke piston seals.

 

I'll do some more logging with rpm input, because vacuum and AFR not enough data though fascinating.

 

 

At mid-range 2500rpm I can hit 10.5 AFR too. Seen 9.8 going steady uphill!

So not far off the Turbo mixture :-) but the lean is scary

 

I guess that is still on the idle jets, maybe main just coming in.

So perhaps my choice of 26mm chokes, but .1 aux venturi is misguided for such a light car and would better suit a heavy 6 cylinder Datsun and I should aim for larger chokes and let it rev.

 

 

 

 

Should I put copper grease or PTFE tape or something on the lambda sensor threads?

 

 

Copper grease might help it to not sieze, but there are better products like this http://www.ebay.co.uk/itm/ANTI-SEIZE-HIGH-TEMPERATURE-THREAD-LUBRICANT-Turbo-Exhaust-assembly-VC13109-/391493767493?_trksid=p2385738.m2548.l4275 Just make sure whatever you use has no silicon as this will affect the sensor

i wouldnt put anything on it , slightest bit of contamination and its finished  

righto... it's probably well sealed with all that lovely carbon. The bung bolt emerged fine after 2 months of use.

 

So I'll adjust the pump stroke and  lower the floats before shelling out on .3 venturis or larger chokes.

 

Edited 20 April, 20179 yr by dvderlm

All setup right. Fuel level at 27mm below too of carb body. http://sideways-technologies.co.uk/forums/index.php/topic/1048-setting-dhla-dellorto-dellorto-float-fuel-level/

 

Access to pump lever threaded rods a bit awkward.Better  disconnect battery as starter motor main terminal fairly close.

 

Suppose lowering fuel float level 0.5mm or 1mm costs nothing.

Connected coil negative to AFR gauge engine speed wire.

 

Wonder if I can get venturis and chokes out in-situ?

 

Edited 22 April, 20179 yr by dvderlm

No I can't get them out in situ. Cylinder 2's venturi had lost its retainer screw and did not want to move. Maybe some dirt in the hole.

 

So took trumpets off - as manifold studs don't give enough clearance when attached. Then took both carbs off .

Adjusted pump stroke for about half duration/volume by screwing nuts anti-clockwise.

Used a length of wood past throttle plate to un-seize the stuck venturi.

 

Took out venturis then chokes, cleaned up. Changed chokes from 26mm to 30mm. 

Throttle plate does not uncover any progression holes.

 

Refitted and drat it races at idle - so maybe air leak as larger choke tube can't have an effect with throttle  butterflies closed.

 

Edited 23 April, 20179 yr by dvderlm

I've removed the 18:1 samples -only get those at idle - gonna reduce timing -, but this is what the change from 26mm chokes to 30mm chokes   with .1 venturi, plus reducing pump stroke,  did to mixture profile.

 

With 26mm - some over-rich 9.5:1...

;

and 30mm - nothing richer than 11:1.  Far more 12.5 to 14 samples for similar journey.  Well a bit more accelerating as the 12 to 12.5:1 peak shows

 

Edited 24 April, 20179 yr by dvderlm

Gauge thinks my engine will do 25999 rpm!

Maybe connection to coil negative is affecting ignition advance too? 

at 18:1 it would be idling pretty rough, have you checked for leaks? if there are none then adjust the idle mixture to ~13.5-14:1

what is your ign timing at idle?

That's the weird thing. It idles really smoothly but now at 1500+ rpm. Will pull the car along if I'm gentle with the clutch.

 

I had it at steady 950rpm about 4 idle jet changes ago (with 36 in .7 holders at 1.25%CO2 and with 48 in .1 holders).

Timing is 2 degrees "centrifugal" advance on 5 static at 1000rpm.(using Aldon Amethyst).

Plus 7 ish degrees vacuum advance at 1000rpm at -7psi and 3 at -3.5psi.

So about 8 to 12 degrees load and temp dependent.

I have a snapshot from this morning at 840rpm with 4 degrees centrifugal and no vacuum so 9 degrees total BTDC.

 

Slight touch of the accelerator pedal even gently and I'm on 13.5:1 pulling the car along and can get to 14 to 15 at cruise at 2500rpm 50mph.

 

My new theory is that the starter circuit is providing the idle fueling. There were no starter jets fitted when I got the carbs.

Perhaps the piston is not blocking flow or the actuator is leaking.

 

It has a weird reduced diameter cutout section on the Turbo carbs.. perhaps it needs a good seal right there... on Turbo car to stop fuel and air blowing out, on N/A to stop

"falschluft" air leaning the mix and activating the starter device circuit. Not taken that bit apart yet.

 

 

I've been screwing in the idle mixture screws as I've gone up on the idle jet getting good vacuum at each carb manometer test point (big pulses of 17inHg).

I may reset mixture with the trusty colortune - wish I had four of them.

I might screw the idles right in or remove the idle jets and see how it behaves. EasyStart spray revealed no air leaks at manifold last time I tried.

There are more places to investigate. The jet stack cover supplies air to the starter device. There's a one way valve to my brake servo and all four runners are connected with a balance tube to measure "manifold"

vacuum. A vacuum gauge on dash shows pulses at idle despite an accumulator chamber and pulse reducer but says I get steady-ish 2to7inHg  at normal loads and 9 to 14inHg on overrun.

The AFR gauge does n't think it goes lower than -8.3psi and drives/cruises at -2 to -5 psi.

 

 

 

Edited 25 April, 20179 yr by dvderlm
17inHg is -8.3 psi

Yes, the Turbo carb, or conversion to turbo use from n/a carb needs an o-ring on the starting device actuator spindle...

https://www.dellorto.co.uk/shop/dellorto-car-carburettors-parts/parts/dhla-parts/dhla-turbo-conversion-kit/

 

If that o-ring has failed there's a leak, maybe on both carbs.

1500 sounds too high for idle

Precisely - air is flowing through to cylinders and dragging some fuel despite the throttle plates being fully closed.

I did have to move the starter device levers to remove the choke cable and I think this has exposed the actuator fault leak.

 

Des Hammill reckons 1200rpm idle is common for twin 40s, but I had got it down to 950 using 36 idles and a very rich .7 holder. Even with this idle the Dignition thought the low vacuum was accelerating and advanced too early, hence switch to mappable ignition - nice and drivable but a bit gutless and capable of more top-end. The AFR gauge showed it was lean at idle at tailpipe (even before it raced) and also helped me find mid-range richness. But I've been told I may need as much as 55 idles on 1300,  55 and .1 was poor economy and 46 with .1 was surging lean, 51 and .1 was poor economy but better than 55. 48 and .1 was okay.  

51 and .5 was brilliant on a cold day.

The Lotus turbo carb also has lots of progression holes so picking the right combination of idle and holder for n/a use can be interesting.

 

 

 

 

  

sounds like you're shafted then mate!

 

jk, but it does sound like you're going down a rabbit hole with this one, esp using turbo carbs

 

get some of these in twin 40 size and be done with it

 

http://www.jenvey.co.uk/products2/heritage/heritage-dcoe-twin-body-45mm-single-tds45

Thanks for the encouragement.

Jenvey's are tempting, but I'm trying to stick to tuning appropriate to the car's age.

 

The 40 in .7 idle holder that 2.2 litre Lotus Esprit Turbo uses begins to make sense if there's a Turbo pushing the air in. And I can always seal up the progression holes to make it like a DHLA40 universal (or E OR C) with only 3 progression holes.

 

A short test with winding in the idle mixture screws to nearly no clearance and it kept running at maybe 650rpm  (hot engine) - not very smooth but I'd expect it to stall, and I then sprayed some short bursts of targeted EasyStart . When applied near carb serving cylinder 1 and 2's jet stack, the engine speeded up - despite starter device being in off position. Didn't do same for the other carb - now it might have been pulled in through the idle air inlet small hole at back of carb, so I'll try again with a rubber tube in that hole. It might be the jet stack cover gasket Might even be the idle air bypass screws for balance. But if it is just a failed o-ring seal on the actuator it's an easy fix.

 

It's only cause I'm trying stupid large jet with leanest holder that mixture screws are open so little. There's more suction on the starter device orifice.

Or I can always run without starting device jets and rely on my preheater and the accelerator pump to get it running in cold weather. 

 

The bypass screws might also be a clue - throttle plate movement is synched and balanced, so variation might be a starter device leak - not any difference in cylinder efficiency such as #1 being nearer radiator inflow or only #4 now having the servo pipe.

 

 

Edited 27 April, 20179 yr by dvderlm

Went back down through the jet combos.Almost definitely a new fault. Won't run on the 36 in .7 holder any more with mixture 5 turns out, though that does need more advance and some choke until engine gone a few hundred yards. Bigger jets don't need choke over about 10 degrees.

Back to 48 in.1 while some Viton o-rings make their way to me

Edited 28 April, 20179 yr by dvderlm

Jeff Hartman's book How to Tune Engine Management Systems has 18:1 as carb lean limit... http://afrplus.com/AFRplus-unit-functionality.asp

---

AFR Values & Characteristics in Four Stroke Engines
 6.0 AFR - Rich Burn Limit (engine fully warm)
 9.0 AFR - Black Smoke / Low Power
11.5 AFR - Best Rich Torque at Wide Open Throttle
12.2 AFR - Safe Best Power at Wide Open Throttle
13.3 AFR - Lean Best Torque
14.6 AFR - Stoichiometric AFR (Stoich)
15.5 AFR - Lean Cruise
16.5 AFR - Usual Best Economy
18.0 AFR - Carbureted Lean Burn Limit <-------- May be okay?
22.0+ AFR - EEC / EFI Lean Burn Limit

Edelbrock Performer guide http://www.edelbrock.com/automotive/misc/tech-center/install/1000/1406_manual.pdf

 

5 AFR RICH BURN LIMIT: Combustion is weak/erratic.
6-9 AFR EXTREMELY RICH: Black smoke and low power.
10-11 AFR VERY RICH: Some supercharged engines run in this range at full power as a means of controlling detonation.
12-13 AFR RICH: Best power A/F: Un-supercharged WOT.
14-15 AFR CHEMICALLY IDEAL: At 14.6 the A/F is at the theoretical ideal ratio with no excess fuel or oxygen after combustion. Good A/F for part throttle cruise and light to moderate acceleration.
16-17 AFR LEAN: Best economy A/F ratio. Borderline for part throttle drivability (worse than borderline if EGR is used).
18-19 AFR VERY LEAN: Usual lean limit (Driveability).
20-25 AFR LEAN BURN LIMIT: Varies with engine and system.
Even though engines will run anywhere between 5 and 25 A/F, the usual target values for an un-supercharged engine are a fairly narrow range (Figure 1). A/F is about 12.5 for the WOT and 14.0-15.5 at part-throttle cruise. An intermediate value of about 13.5-14.0 is usually used for mid-range power (non-WOT acceleration).

Will finally run on 36 in .7 holder again with mixture screws about 6.5 turns out and a bit more advance. Remember the DHLA tower screw is very fine threaded. So any Weber DCOE advice about 1.5 to 2.5 turns is irrelevant.

 

Engine idle a bit lumpy though. Vacuum loss every so often like an inverse reggae drum beat 1-3-blank-2: Reason: cylinder 4's idle bypass screw was open a little bit as well as cylinder 3 quite a lot. Using an infrared thermometer I could see number 4 exhaust manifold pipe was only at 80 degrees while the others nearer 230. Closing number 4 bypass and opening mixture a bit more and cylinder 4 joined in the game. Used a vacuum gauge to set it for highest vac. That smoothed it.. and I could set sensible idle speed....but still 18:1 idle.

 

I then actually managed to get 40 in .1 to take me round the block, with mixture screws in a fraction but need to increase the pump stroke - so awkward with carbs on the car and engine hot.

But

 

Need to repeat the Easy Start rev test with these jets.

 

New theory: 

The idle bypass adjustment to achieve cross cylinder balance  (and road-going economy) is the real reason for 18:1 only at idle?

 

 

 

No enlightenment from the Easy Start.

 

So closed the bypass screws like a race car. Still 18:1 and idles at 2000 when hot.

 

So moved to 38 in .7 holder and got mostly 18:1 but sometimes 17.3:1 at idle.

 

So reset ignition timing to 8-4-12-17-27-32 centrifugal with much more vac advance.

Backed off speed screw completely. Screwed in mixture screws and idle speed back down to 900ish but not quite smooth as a little unbalanced.

 

Need to put the pump stroke back up a bit  again as this gets a lean bog accelerating from 1500 to 2200.

 

But once over 2600. .. then it's fun again. I think it's reading  to be 13.2:1 at W.O.T. with 110 main and 185 corrector in 30 chokes.

The gauge needs a microphone input so it can log when I press the loud pedal. Even ignoring the 25,999rpm anomalies I think the ignition advance confuses the square wave  sampling in my AFR gauge so I can see revs rise but  peak 12,243 rpm is obviously not correct or helpful.

Maybe that was  6100 rpm? 

 

Have ordered a Hall sender tacho kit to try measuring crank pulley speed with a magnet.

 

This is mixture profile of 5.1 seconds of hard acceleration. Can even see the manifold go from vacuum to +0.3 psi pressure from tuned trumpet ram effect.

 

 

 

Edited 4 May, 20179 yr by dvderlm

Put the pump stroke back up. Just disconnected battery earth*, removed trumpets, undid 8 Thackray washer mounting bolts and fuel inlet, left accelerator cables and starting device cables attached. Lifted carbs onto firewall, applied 6mm? nut driver and plier to expose more threads.

Tested that it would squirt two straight jets of petrol about 4 feet from each carb over the engine bay and onto the driveway!

 

Refitted. Drove to Wales and back averaging ~60mph (200 miles in 3hours 20 via M6 spaghetti junction and A-roads) and consumed 50.7mpg. It's very happy cruising at 3200+rpm and gets 14.7:1 to 17.5:1 AFR.

and despite some outside lane fast sections did not go below 13.5:1.

 

This is a 90 minutes of that journey. If it wasn't for the 18:1 readings when I take my foot off the gas pedal I'd be delighted. 

90% of the time it was not too lean... but I can't stop fuel flow like an injection car. Either unburned fuel due to cam overlap

with poor filling misleading the lambda sensor, or an airleak.

 

So I've ordered some DHLA Turbo spindle seals, o-rings  and star washers to fit along with the starter device actuator viton o-rings when they arrive.

 

* I disconnected the battery because I had tried to temporarily  increase the pump stroke by wrapping wire around its rod. I have a wonderful burn line on my finger from the wire touching the starter terminal. 

 

 

Edited 13 May, 20179 yr by dvderlm