AEM UEGO AFR sensor on a B50 with JRC PWK

Here you can talk about all kind of TECH for the B50 B44 B25 And Other BSA unit Singles

Moderator: minetymenace

Post Reply
bbenxe
>80
>80
Posts: 81
Joined: Tue Apr 11, 2017 12:00 pm
Location: Bristol, UK

AEM UEGO AFR sensor on a B50 with JRC PWK

Post by bbenxe »

2 years ago I snapped the con-rod on my B50T, and packed it off to Roger Taylor in Gloucester for a rebuild. It came back packed with goodies, of course including a new NEB con-rod, but also with his filter conversion, MX cam, and a stub conversion with light porting to fit a modern carb. I had hoped to put on a TM pumper carb but settled on a 32mm JRC Keihin PWK licenced copy from BrillBikeParts given the space constraints against the frame.

Other modifications on the bike include a reed valve primary breather, and long stanchions for the X75 Triumph with Racetech cartridge emulators and progressive springs hidden inside, plus NJB shocks to lift the bike for more ground clearance.
20190831_112024.jpg

The carb worked OK out the box with Brill's recommendations as below...
jets.gif

But despite all our adventures in the next 2 years the carburation has always felt a bit rich. I use the bike to ride the few miles to work, bar-hop, greenlane around Bristol, Wales and Somerset, and take my girlfriend for sunny cruises. Rare 8-stroking, occasional plug fouling would sometimes spoil these, and I couldn't fix it despite moving the needle to clip 1, and the pilot was already the leanest available, 35.
IMG-20200927-WA0007.jpg
IMG_20201025_152832_236.jpg
IMG_20201231_145459_139.jpg

So this year I decided to fit a wideband sensor in the exhaust to sort the carburation once and for all. I knew a new needle profile was needed and this seemed to be the best way to figure it out. I've some experience setting up a PWK previously for my racing Bultaco, and didn't fancy the pain again or repeated expense of dyno testing.
I ordered a wiieband 'UEGO' (Universal Exhaust Gas Oxygen) sensor from AEM, and had the supplied bung welded into the header 18" downstream of the exhaust valve. I also attached an exhaust clamp and packed it with silicone in an attempt to prevent air leaks at the head.
IMG-20210225-WA0003.jpeg
IMG-20210427-WA0014.jpeg
IMG-20210427-WA0007.jpeg

That was the easy part... Wiring it proved extremely difficult. Running the sensor and gauge off a remote battery worked fine. It would also work fine plugged into the bike's circuit up to the point of starting the bike. A few seconds in and the gauge would go haywire, repeatedly shutting down and flashing nonsensically. I suspected excess voltage despite my modern reg/rec, and a digita gauge across the battery would vary wildly but include readings of >16V, the rated limit of the gauge. I attempted to use a capacitor across the battery to 'smooth' the voltage (my electrical ability is mostly guesswork) but the solution I have settled on is using the battery to power the gauge and sensor alone, running total loss. Using a capacitor in place of the battery for the rest of the bike's circuit is working well, despite having a Wassel Vape EI which I wasn't sure how well would cope without a battery.

With the gauge finally working I set out on a quick ride around town trying to get some readings before the rain and 50mph winds hit this very damp Bank Holiday Monday in the UK.

Idling gave a slightly lean reading of 14-14.5:1. 13-14:1 would be preferable.
Pulling from 1/8 to 1/2 throttle would give very rich readings of 9-10:1, when 12:1 is the recommended for power.
I didn't spend much time beyond 1/2 throttle before the rain scared me home as the AEM is meant to be sensitive to water ingress, but was also seeing rich readings there
IMG-20210427-WA0016.jpeg

With some useful information I returned home to play with the incredibly useful tool at Kyajet.de I could see that with a JJL and JJN needle instead of the standard JJH I could play with different clips and main jets to lean out the appropriate throttle positions by the ~20% required to get back to the correct AFR.
Screenshot_4.jpg

I believe JJN to be the same needle Gunnag used with a JRC PWK so expect it to be much closer to the correct setting. Needles are in the post now so I will report back when I've been able to test. I'm hoping for more reliablty, more power, and more smoothness!

Might be a few weeks though as this weekend is the first round of the UK flat track series for 2021 in which I race my Bultaco. I came second in the Vintage two stroke class last year (mainly by dint of there only being one un-cancelled round and me having a very good day that day...) but I have high hopes this year so am fastidiously preparing.
Last edited by bbenxe on Tue May 04, 2021 11:57 am, edited 1 time in total.
'71 BSA B50 street scrambler
'75 Bultaco 250 flat tracker racing as #404 in the UK DTRA National and European series
'69 East Coast export A65 Lightning
baz
>280
>280
Posts: 297
Joined: Mon Jul 10, 2006 7:17 pm
Location: ESSEX

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by baz »

Your AFR sensor looks a lot better than my narrowband Koso
I used mine to great effect on my Norton commando, although I couldn't get exactly the best reading it did get me very close and the bike running much better
I marked off the throttle drum and made a note of how it ran on each circuit and this gave me the correct pilot setting, changed my slide cutaway from 3 to 3.5 dropped my needle one notch and came down on the main jet
bbenxe
>80
>80
Posts: 81
Joined: Tue Apr 11, 2017 12:00 pm
Location: Bristol, UK

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by bbenxe »

baz wrote: Mon May 03, 2021 6:47 pm I marked off the throttle drum and made a note of how it ran on each circuit
Yes this is absolutely essential, I've done the same on mine. It surprised me how much time is spent at < 1/4 throttle which emphasises the importance of getting this section right!
'71 BSA B50 street scrambler
'75 Bultaco 250 flat tracker racing as #404 in the UK DTRA National and European series
'69 East Coast export A65 Lightning
User avatar
SteveS
>640
>640
Posts: 640
Joined: Mon Jul 10, 2006 8:28 pm
Location: Nuneaton, Warks. U.K.

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by SteveS »

Thanks for sharing :thumb

I will be interested in your results with the new needles
Steve Sewell
B44Claus
>220
>220
Posts: 224
Joined: Fri Jul 07, 2006 12:49 pm
Location: Denmark

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by B44Claus »

Hi Ben,

Very interesting tests you are doing with the PWK carburetor <200. Is your carb 30 mm as the table from Brill Bike Parts suggests?
I have a 34 mm PWK(OKO) carb I'll try on my B50/B44 Hybrid build and will therefore follow your findings closely.

Thanks for sharing!

Claus
B44Claus
bbenxe
>80
>80
Posts: 81
Joined: Tue Apr 11, 2017 12:00 pm
Location: Bristol, UK

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by bbenxe »

B44Claus wrote: Tue May 04, 2021 9:14 am Is your carb 30 mm as the table from Brill Bike Parts suggests?
Good point Claus, I should have mentioned this, so have edited the original post to say so. My carb is the JRC 32mm.

The Keihin PWK comes in two variants which take two needle profiles:

28mm takes N427-46, which is 57mm long, 2.6mm diameter.

33,35,36,38,39,41mm take N427-48, which is 66mm long, 2.9mm diameter.

A JJH N427-46 (2.435mm initial diameter) is very different to a JJH N427-48 (2.675mm initial diameter), so be sure to get the right needle type!!

The needle crudely marked JJH in my 32mm JRC PWK was 61mm rather than 57mm long, but otherwise matches the spec of Keihin's N427-46-JJH.

The external dimensions of the JRC 32mm are in-between the Keihin 28mm and 33mm, it would appear to be their own shape casting, but given the needle measurements best to use parts for a Keihin 28mm if swapping out jets/needle for Keihin originals.

The danger of using N427-46 in the JRC 32mm could be that at 57mm long it doesn't stay in the needle jet at full throttle (which could be why the carb came with a 61mm long needle that doesn't match a Keihin needle exactly??) and so could jam the throttle fully open if it did so.

I'll test this as I install and report back.
'71 BSA B50 street scrambler
'75 Bultaco 250 flat tracker racing as #404 in the UK DTRA National and European series
'69 East Coast export A65 Lightning
User avatar
kommando
>1400
>1400
Posts: 2061
Joined: Mon May 24, 2004 8:41 pm
Location: Scotland

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by kommando »

The ignition emf will pay havoc with a digital meter and no doubt the AFR sensor, an analogue meter will tell you the real voltage and if it is still peaking at 16V then your rectifier, like the 3 phase Sparks one was, could be affected by the emf too. Either resistor plug caps, HT lead or resistor plug will possibly fix or AC could be leaking out into the DC circuit.
Jeff K
>1400
>1400
Posts: 1444
Joined: Sun Jul 13, 2014 12:30 pm
Location: Bay City, Michigan USA

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by Jeff K »

As Kommando has pointed out you are using a very sensitive piece of equipment that likes a stable power supply. You are also using it very close to a steady supply of stray signals. You could also try a grounded shielding cover on the meters wires, like what is used on a TV antenna's Coax cable. I have a large hand held automotive analog Oxygen ration meter that I use a lot on Motorcycles. You can see what is happening as you slowly open up the throttle and go through the different jets and ranges. You would be surprised how many times that you think it was lean only to find that you had a section that was going a little rich. I had the old head mechanic at a long gone local Triumph shop tell me decades ago that 90% of Carburetor problems are electrical. I still laugh about how often is is still right on that one.
Thanks for posting this.
Jeff
bbenxe
>80
>80
Posts: 81
Joined: Tue Apr 11, 2017 12:00 pm
Location: Bristol, UK

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by bbenxe »

Jeff K wrote: Tue May 04, 2021 1:01 pm As Kommando has pointed out you are using a very sensitive piece of equipment that likes a stable power supply. You are also using it very close to a steady supply of stray signals. You could also try a grounded shielding cover on the meters wires, like what is used on a TV antenna's Coax cable.
Yes, I spent quite a time moving wires around to see if having them near coil or HT lead triggered the problem but the problem does seem to just be whether the gauge is in same circuit as the rest of the bike rather than a remote battery
[/quote]
kommando wrote: Tue May 04, 2021 11:57 am The ignition emf will pay havoc with a digital meter and no doubt the AFR sensor, an analogue meter will tell you the real voltage
Will keep an eye out for one, I suspected this might be an issue as I had read before that most consumer digital gauges struggle with non-static (dynamic?) readings so didn't want to read too deeply into the voltage measurements I was seeing.
kommando wrote: Tue May 04, 2021 11:57 am Either resistor plug caps, HT lead or resistor plug will possibly fix or AC could be leaking out into the DC circuit.
Good idea, will try these. In the meantime I'm happy running it off the battery and rest of the bike off a capacitor (which means I have to remember to put the battery back on charge when I get home) as it's not intended to be a long term solution. I want to use the sensor and gauge to get the jetting right, and then remove the lot. Expensive moisture-sensitive electronics aren't compatible with a go-anywhere mud-plugger, plus it spoils the classic look :cool:
'71 BSA B50 street scrambler
'75 Bultaco 250 flat tracker racing as #404 in the UK DTRA National and European series
'69 East Coast export A65 Lightning
User avatar
JB
>500
>500
Posts: 513
Joined: Tue Jun 15, 2004 8:32 am
Location: Devon, England

Re: AEM UEGO AFR sensor on a B50 with JRC PWK

Post by JB »

I fitted the exact same sensor/gauge to my B25T when I built in in 2017, it worked right from the off without any problems and that was with the luddite zenor with a solid state rectifier and an MB9U motobatt fitted.
Admittedly my bike is running an AMAl, so my initial settings weren't far off, but as I built the bike for trailriding I wanted the AMAL as good as it could be and it paid off brilliantly, it now carburates really cleanly all the way through. What I found was that the settings I ended up with were all on the lean side compared to the std settings! I'm not sure whether that was down to running the engine with modern guides with much tighter stem clearances or modern fuel... probably both! I also ended up with a bike with really low fuel consumption, I often go out trailriding with others on modern 4 strokes and find it funny that I can skip every other fuel stop when they're all fretting about running out :lol:
You're right to be weary about getting the gauge wet, my gauge was also only intended to be used short term, even then some water got in thanks to the useless high level front mudguide(guard :laugh ) and stopped working.
Post Reply