Using Twin Lambda sensors - anyone else tried this ?

Technical advice Q&A
User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#1 Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Fri May 26, 2017 9:56 pm

http://www.webcon.co.uk/shopdisplayprod ... d+Displays

Having used Webcon's AFS-1 Lambda read-out to good effect on the Mini, I have now gone further with the set-up on the Jag.

Image

Webcon have two set-ups like this, I use the AFS-1 (ostensibly used on motorcycles) as it's smaller ; technically the systems are identical.

They use lambda-sensors of a type that are standard fitment on closed-loop fuel-injected cars, so give an instantaneous read-out that is sensetive enough to rattle the LEDs when you go over a bump in the road on a float-chambered motorcycle anyway.

I had initially intended to use the same read-out panel and switch it between the two cars, having even made a clip-housing on the Mini dash for easy removal, but I'm so taken with the on-board Game-Boy that I've bought a second one and have left the first one in the Mini as a permanent fixture.

Image

Image

This post is a little premature in that the car is getting a new interior at the same time, which is why I've been able to incorporate all the wiring, switches and read-out location modifications, so I've not used the system on the Jag yet.

However I'm confident that I'm on the right track and thought others might like to see how it's gone together.

The essential twist on the Jag is that we have to sniff two engines, one bank composed of cylinders 1-2-3 and the other 4-5-6, each running one-and-a-half carburettors, and this requires two lambda sensors - one in each exhaust collector.

And you then either need two read-out panels to follow both banks at once, or one read-out that you can switch between banks, which is what I'm doing.

I put the tapped bosses into the exhaust pipes a few years ago in preparation for this, it was relatively easy to find adjacent spots where there was space for the sensors and - above all - space to get at them with a spanner etc.

Image

The wiring comes into the cockpit near the windscreen-wiper bottle on my RHD Series 1, via a spare grommet, and I chose to locate the Superseal-type plugs on the lambda sensors out of sight down by the clutch bell-housing.

Image

Image

Image

I re-wired the AFS-1 with much thinner wires taken out of some RG45-type cable as the read-out panel only uses tiny currents to drive the LEDs, the original cable choice being dictated by robustness - whereas I needed to take the wires out through the top instead of the bottom of the instrument and this required some dexterity.

Image

I used stainless piano-hinge to carry the LED-panel in the roof of the glove box, with a couple of carefully positioned centre-point blows to render it stiffer, and it stays put out of the way and almost out of sight when not required.

Image

Image

Image

The lambda sensors use 4 wires, two being a simple heating circuit carrying a couple of amps, and the other two carrying the signal. The trick is to power both lambda heating-circuits at the same time, so the read-outs are permanently available, but switch the read-outs so that you can flick between cylinders 1-2-3 and 4-5-6.

Image

I used one switch to act as on-off for the whole system, and a second switch to jump the read-outs from one set of cylinders to the other - and then hid both of them under the under-dash panel on the passenger-side of the car (made easier by having made new steel panels for all the under-dash area).

Image

Image

There is actually VERY little room for switches and wiring in this area, and in usual Jaguar fashion you have to be VERY careful not to overlook what-is-now-going-to foul-what-now-that-I've-modified-something.

Image

But for once I didn't shoot myself in the foot and think I got it right first time.

I believe this will be a useful tool to improve the choice of SU needles for the car, just as it has proved to be on the Mini, and it's sufficiently discreet to be able to leave it permanently installed.

As in, Now You See It .......

Image

........ Now You Don't.

Image
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#2 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Sat May 27, 2017 12:28 am

Clever stuff - I like this. I'm thinking of doing something similar to my Harley in Florida, to which I've just fitted a 120" (2 litre) motor, and I think you've shown me the way forward.
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links


christopher storey
Posts: 5698
Joined: Sun Mar 09, 2008 3:07 pm
Location: cheshire , england
Great Britain

#3 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by christopher storey » Sat May 27, 2017 8:22 am

There is one aspect of the readings which needs always to be considered : because of the balance passages between carburetters , all 3 carbs in fact affect the readings in either pipe, although the end carb's effect on the "opposite" pipe is less than that of the other two. I have found that when setting up, ( in my case with an infra-red CO meter ) it is best to optimise the centre carb before attending to the outer ones and then returning to the centre carb for final tweaking

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Heuer
Administrator
Posts: 15174
Joined: Sat Mar 01, 2008 5:29 pm
Location: Nottinghamshire
Great Britain

#4 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Heuer » Sat May 27, 2017 9:57 am

I have a Lambda sensor that attaches (temporarily) to the exhaust pipe tip with a portable meter that can sit in the cabin. Although the exercise was interesting it added little to the overall tune of the car because of the 3/2 factor. If I were to do it again I would fit three sensors, one in each of the carbs exhaust manifolds 1, 3 and 5. This would allow the SU's to be set up and constantly monitored if required with a rotary four position switch (Off, 1, 2, 3 sensor). Doing that with cast manifolds would be difficult unless you could machine a screw hole in one of each pair but would be easy with tubular manifolds.

If you are going to this much trouble Rory you seriously need to consider EDIS/Megajolt ignition and also a Road-Dyno if you can find one. The EDIS will give you the largest performance gain of all with twin maps and a display you can see on a laptop whilst on the move. The Road-Dyno gives you torque and power curves so you can see what improvements actually work.
David Jones
S1 OTS OSB
1997 Porsche 911 Guards Red
2024 Lexus LBX

Add your E-Type to our World Map: http://forum.etypeuk.com/viewtopic.php?f=1&t=1810

Link:
BBcode:
HTML:
Hide post links
Show post links


JagWaugh
Posts: 559
Joined: Sat Sep 21, 2013 12:56 pm
Location: Eglisau, Switzerland
Switzerland

#5 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by JagWaugh » Sat May 27, 2017 10:32 am

Someone's been busy....

I'm with David, you're better off with one sensor/"Carb-branch", but it is a bit difficult on a CI manifold.

If I were going to do it on a CI manifold then I would probably:

Weld up a "Land" for the plug with Si-Bronze
Mill the land flat
Drill and tap for the sensors

But that is a fair bit of bother. Less than making a set of tubular headers from scratch, and more than TIGing a bung into a set of bought tubular headers.

But if you've a set of CI headers you don't mind sacrificing, some Si weld filler, a milling machine, and a day or 3 to spare...

Link:
BBcode:
HTML:
Hide post links
Show post links


Annivar
Posts: 15
Joined: Fri Dec 30, 2011 4:32 pm
Location: Windsor Canada
Canada

#6 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Annivar » Sun Jun 04, 2017 5:16 pm

Yes, I just installed the Innovate dual lambda kit in my series 1 FHC.Image
This includes 2 O2 sensors and a dual gauge that can be toggled to display in A/F ratio or lambda. The impetus for this was that I noticed last summer that when I thought I had the correct mixture setting at idle, the plugs looked quite lean after a highway run. By richening it up at idle I had a better plug colour at speed but a fairly rich lumpy idle. I planned to try different needles that were slightly richer mid range but wanted to be able to more objectively measure the differences. When I saw this dual lambda kit I thought I would give it a try. I welded bungs into the exhaust and ran the wires up each side of the floor and firewall.
They enter the cabin through the small blanking plate. I chose to make a new plate with a hole for a grommet and the original is untouched should the system ever be removed. The only permanent change to the car is the hole in the glovebox for the wires which are all tucked in behind Image
Image
Imagehttp://i1106.photobucket.com/albums/h37 ... novate.jpg[/img][/url]

Attached is a short video of the gauges on initial startup. As you can see the choke is still on. Ideal A/F ratio is 14.7/1 so these numbers indicate a slightly rich mixture as expected. I have since done some tuning and driving and am very happy with the results so far. I am surprised how reactive the sensors are to turning the jet adjusting screw. The slightest turn (1/16) or less has a significant effect on the readouts.The gauge can be hooked up to a PC or data logger for recording and a more extensive analysis of the data but I haven't done that yet. I still haven't tried the new needles but am confident that I will obtain objective data. Hope this helps. Thanks

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#7 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Sun Jun 04, 2017 6:46 pm

Very interesting - thank you.
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#8 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Mon Jul 10, 2017 9:51 pm

Missed your post Annivar, or would have replied earlier - thanks for the input.

I might have gone for the same system as you had I known a twin read-out existed already, although I have to say the "analog" LED-curve might be easier for me to interpret than two sets of figures - and anyway I want to remain with the same stuff across all my vehicles so I'm a bit tied to the AFS meters.

A recent 1.000km + trip across Europe has led me to conclude that what I have done is considerably more useful than I had imagined, and perhaps been given credit for here.

The two read-outs can be flipped from 123 to 456 instantly, and I am apparently working on a set of carbs where the current settings are giving me exactly the same carburetion across both banks - to within an LED, which is best the eye can register.

This COULD mean that carbs 1 and 3 are (for example) horribly rich, and the centre #2 is terribly weak, to get identical readings, but since the set-up is recent and the jets and needles new, I'd guess that this isn't the case.

What is very apparent is that with the current needles (I can't remember what they are but I know they are THE ones recommended and sold everywhere) I am getting perfect combustion at tick-over, which is where - of course - we all set things up, but that the carburation is progressively weaker and weaker as I go faster.

So much so that I didn't dare go much over 90 mph as I was well into the "red" LEDs.

This probably explains why I have always experienced some pinking under acceleration ..........

I'm reluctant to "just stick it on a dyno" as I believe this bears no relation to cruising at 2.500 rpm across France, which is all I'm interested in.

How do I relate my Lambda readings to SU-piston position without putting a camera under the bonnet, calibrated extensions on the piston-damper rods, and watching which of the sixteen diameters (?) on the SU needle I'm using, please ?
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#9 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Mon Jul 10, 2017 10:56 pm

I take it you are aware of the 'Minty Lamb' graphs for SU needles? They are a good guide for starters. I THINK the piston position will pretty much correspond with pedal position - in other words with the amount of air the engine is drawing. That makes sense to me - although I've never checked it.
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#10 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Thu Aug 03, 2017 3:30 pm

Update on the needles / lambdas, after the Mediterranean-to-Cape Wrath-and-back expedition.

I could have sworn I had fitted UE's about 15 years ago, with new jets, but discovered they were the correct/original UM's - which Mike Wilkinson says is all they ever use on cars that aren't heavily modified.

So is the very-weak running I'm getting being provoked by the wrong throttle-piston springs, perhaps ?

No trace of any colour-coding on the old ones, and not sure I trust my scales, so now waiting for a set of 3 new ones from Burlen ...........
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links


MarekH
Posts: 1752
Joined: Wed Jul 21, 2010 8:30 pm
Location: Surrey
Great Britain

#11 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by MarekH » Thu Aug 03, 2017 6:23 pm

Dear Rory,

I use a similar but different setup on my v12. currently I have four bungs welded in at the downpipes and two in place of the crossover pipe.

I think it will be problematic to put them into the headers as David suggests, as they will overheat. The Innovate LC1 manual doesn't suggest a minimum distance but does state that the temperature at the bung mustn't exceed 500'c.

One thing which will help tuning (or diagnosis of differences) per carburettor (or per cylinder) is exhaust gas temperature measurement. This can be done in the headers or even closer to the exhaust valves.

See http://www.jag-lovers.org/snaps/snap_vi ... 1413043806

kind regards
Marek

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

juicerider
Posts: 60
Joined: Tue Oct 20, 2015 1:15 am
Location: Perth WA
Australia

#12 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by juicerider » Fri Aug 04, 2017 2:40 am

Yeah I've used a single one on mine to tune up a set of webers. I only used one as with webers the same jets go in all barrels anyway.
I originally had the carbs tuned on a rolling road, however I was never satisfied with the tune. It was OK on the main jets but I discovered that driving at a constant fast freeway spead there is very little load on the engine and the carbs were still running on the idle jets. Very useful in determining real world tune rather than just max power tune.

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#13 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Fri Aug 04, 2017 5:58 am

Indeed. How many horse-power you can squeeze out of an engine at 6,000 rpm is irrelevant for most drivers most of the time.
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#14 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Sat Aug 05, 2017 1:01 pm

Exactly my thinking Hugo, and borne out ever more clearly on my recent Road Trip where the staggering flexibility of the 3.8 was the Number 1 factor in making it a joy to drive.

Plus, when you do cane it a bit between 3 and 4.000, it thrives on it, and yet will footle along at 900 rpm in 4th on a whisker of throttle.

My pinking and lean-ness issues spoiled the enjoyment somewhat, however, but for once I'm going to try and sort these BEFORE I blindly rip everything apart and then end up wondering where the problem came from.

See upcoming separate posting, be very interested to hear experienced inputs.
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#15 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Sat Aug 05, 2017 1:25 pm

Pinking and leanness is probably one issue not two; get the mixture right & the pinking will go away, or at least be alleviated.
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Heuer
Administrator
Posts: 15174
Joined: Sat Mar 01, 2008 5:29 pm
Location: Nottinghamshire
Great Britain

#16 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Heuer » Sat Aug 05, 2017 2:16 pm

Rory - I assume you ARE using Super Unleaded?
David Jones
S1 OTS OSB
1997 Porsche 911 Guards Red
2024 Lexus LBX

Add your E-Type to our World Map: http://forum.etypeuk.com/viewtopic.php?f=1&t=1810

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#17 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Sat Aug 05, 2017 4:21 pm

Sure, Shell V-Power or BP Ultimate only, when available.
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Hugo
Posts: 1264
Joined: Mon Aug 29, 2016 11:16 am
Location: Horsham West Sussex
Contact:
Great Britain

#18 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Hugo » Sat Aug 05, 2017 4:47 pm

My local filling station in Florida sells ethanol-free :P
Hugo Miller - rebuilding an imported Series II OTS & converting to RHD

Link:
BBcode:
HTML:
Hide post links
Show post links

User avatar

Topic author
rfs1957
Posts: 1593
Joined: Sat Jan 15, 2011 2:52 pm
Location: Languedoc - France
France

#19 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by rfs1957 » Sun Oct 08, 2017 8:05 pm

Check out the parallel posts that update the information gleaned so far.

viewtopic.php?f=3&t=11056#p89228

viewtopic.php?f=3&t=10464
Rory
3.8 OTS S1 Opalescent Silver Grey - built May 28th 1962

Link:
BBcode:
HTML:
Hide post links
Show post links


Woolfi
Posts: 315
Joined: Thu Jan 28, 2010 5:57 pm
Location: Germany
Germany

#20 Re: Using Twin Lambda sensors - anyone else tried this ?

Post by Woolfi » Thu Oct 12, 2017 3:24 pm

It is important that the Lambda measuring system is using wide-band and NOT narrow band sensors. In a normal modern car are narrow-band lambda sensors mounted. Also the electronic must be for the wide-band or a different electronic for the narrow-band sensors. But to measure lambdas from 0,75 to 1,2 a narrow band sensor is not good.
Innovate is using wide-band sensors plus the correct electronic.
The mixture of a perfect tuned car shall be:
1. Lambda 0,9- 0,95 at iddle. Otherwise the motor has some missing ignitions. Fat mixture.
2. Lambda 1,05 - 1,15 at part throttle / crusing, up to 70% of max power. Lean mixture.
3. Lambda 0.85 to 0,9 at more than 70% of maximum power. Fat mixture.
Lot of people think that at part throttle a mixture of Lambda 1,05 - 1,15 is too lean. Pistons can burn. This is WRONG ! ! ! The exhaust gas temperature at part throttle is too low, that pistons can burn.
I am driving my EV12 with a 6,0 l motor since 6 years with a lean mixture at part throttle. Nothing bad has happend with the motor. The mileage is good. Normal cruising with roundabout 20 mpg (5th gear and a 2,88 axle).
15 years ago, when I started to fiddle with the carbs of my car, I have bought a book about carbs on a flee market. This book was from a university of former East-Germany (DDR). There a technical draws which shows very exactly, how the mixture shall be for professional produced car in the senties.
Regards Wolfgang Gatza
At part throttle there is no need for a lean mixture. The production of unburned HC at a lean micture is lower that at roundabout Lambda 1 with a carbed motor.

Link:
BBcode:
HTML:
Hide post links
Show post links

Post Reply Previous topicNext topic