AFR Sample Chambers - Take 3 - R
Turbo Mini Forums

this is all very interesting stuff and good to see some progresion.

like others on here i have just one wideband sensor in the downpipe, it would be interesting to know how the readings from your 2 sensors compares to the reading from one in the downpipe. would it be as expected half way between your two readings? or as there are two outer ports would the downpipe reading be nearer the outer port reading?

I do actually have a third one available which I will (eventually) fit in the bottom of the downpipe. Unfortunately it is a different make (Innovate) which means I may not be able to get identical calibrations as you can only calibrate (easily) at one end of the range, ie free air.

IMO the reading should be exactly half way, ie, the average. The differences I think I can see on Paul and my graph's on the outer is a slightly lower spike, ie a slightly sluggish response to a transient (such as blipping the throttle) but for 99% of the time the readings are following each other accurately - ie the difference between the two readings is constant everywhere else.

The small difference in the fast spike I think is because the outer sample is half the size so comes down the sample pipe a bit slower.

But looking at the graph, I really don't think it is an issue.

BUT, the point is to use the siamese code to get the inner and outer AFRs the same anyway, so then the combined flow must be the same again....

Schrödinger's cat - so which one am I ???

thats intersting rod , the air fuel imbalance flips at a higher rpm ,and the outers get rich .it would be very intersting to see and drive log of that setup ,with a single sensor in the down pipe as well ,but i guess your motivated to get boosty wiv it ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Robert, I've seen the AFRs flip even at low RPM, just using a standard carb and twin widebands on the N/A MG twin downpipe manifold, just from going from on choke to off choke.

BUT, everything I have done so far has been on engines sitting on the garage floor - no load...

So my real motivation at the moment is to ignore all the electronics and fuel injection which I now have working, and get back to some basic cutting/grinding/welding..... so as to be able to put the engines under some load dragging a shell around.

So, after a good weekend away (beer festival up in Norfolk), I shall start on the shell tomorrow.

Schrödinger's cat - so which one am I ???

an old thread i know..
looking to find the sweet spot for sizes of the chambers and feed lines?

Not enough users/data to identify the "sweet spot".

I've got good data from 4mm ID, 6mm OD feed into a piece of 1" nominal bore tube and a 16mm outlet pipe.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I do like that manifold setup

If something is worth doing, it's worth doing half of.

On 28th Nov, 2013 John said:
I do like that manifold setup

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I'll take it off your hands and replace it with a nice wedge of wood in that case

If something is worth doing, it's worth doing half of.

As Paul says, "Not enough users/data to identify the "sweet spot"" yet.

From the few photos of mine remaining after my Photobucket account screwed itself, you will see the key dimensions are pretty much the same - ie, I copied Paul.

I'll see if I can re-install my other photos later (Photobucket doesn't seem to assign new file names so the links appear to persist if the photo is re-loaded).

The other one I'm aware of is Graham's (Graham T on here), again, similar dimensions as I seem to remember posting him some offcuts of my tubing. His difference is he has run from both outer ports and Tee'd them together. Again, limited data to make a comparison.

Schrödinger's cat - so which one am I ???

Of course, the problem is that the exhaust gas going to the sample chambers is bypassing the turbo. This could slightly raise the rpm at which the turbo starts producing boost. It's not really an issue once on boost as the wastegate is open anyway.

You could go smaller on the bore of the sample pipes, but you could then introduce more lag in the readings. At 4mm bore, the lag in indiscernible. Hence it may be worth trying 3mm.

Bigger sample pipes and you will just bypass more and increase the spool-up time.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

ok first of thanks guys for helping share your work and outcomes this is why the forum is so good!

i think ill try a 3-4mm feed line as said to try help spool up i figure i can change it if its to slow on the readings..
i think i rember some talk about the angle at witch we dump the gas back into the down pipe? pauls one looks good to me due to the angle with the flow? if that adds up?
i might put 3 flange nipples on the mani and to the "T" join thing so as to possibly equalize the flow to each o2? clearly u cant get the 100% match on cirstances from inner to outter but we can get close im sure..

ive got monday off so will see to gettign some supliys and have a crack and makeing some up

this is probaly not needed for the hiff44 but if i can use the data to make my needle and stay safe ill be a happy chap
it will be much more apriate when i do the efi later on

I did receive some negative comment for this:



However it did thousands of miles without issue.

Don't worry about equalising the sensor flow - it just needs a sniff.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

ok so ive gotten my materials

it probaly dosent matter but ill ask any how..

should i have the feed line enter the chamber slightly offset so the gas entering the chamber kind of swirls around the end of the O2 sensor before dumping out the down pipe?
or
have it enter square with the gas dividing and moveing around the O2 then down the down pipe?

well ive made the chambers yet to put the feed fitting in and attach them to the mani tho

my main problem is space as the whole thing with sensor is quite long and i need it in a place that allows me to get the mani on and off with out interference or at leas the least

im thinking i may need to cut and weld to make an angle so as to fit them in parallel ish with the down pipe yet feed in at the angle if that makes sence?

ive ended up with a 3.5mm id feed line, 25mm id chamber also 25mm long and the dump tube is 15mm id

the O2 ends flush with its threaded part
ive got some brass compression fittings for the ss feed lines as well

so at this point should i weld the feed line bug so its in the center of the chamber or bias it to the top near the O2?

I would do it as the last photo.

As for swirl, I would avoid it if you can. It has a habit of separating and creating extra losses.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

^^^ as above.

I only put my sample tubes in at an angle as Jean had pointed out the shape of my tubes might collect condensation and spray it direct at the sensor's critical bits when starting the engine.

As for the delay with sample tube sizes etc, I have a third wideband in the exhaust after the turbo and during fast changes in AFR, there is sod all difference in the response times, 1/10sec at the most.

EDIT - I would also suggest stainless compression fittings, not brass - the brass (probably mostly copper) rings simply won't bite into stainless tubing and the brass fittings themselves could easily fail at high temperatures.

Edited by Rod S on 30th Nov, 2013.

Schrödinger's cat - so which one am I ???

will do im off to finish them

right this is what ive got so far

errr... edit for pics...





im not sure how much heat the turbo blanket can take from the outside?

also need to fit the feed lines and stay away from the master cylinders as much i can

Edited by Turbo This.. on 1st Dec, 2013.

Looking good....

I've reloaded all the "missing" photos of mine back into photobucket so if you go back to page 1 of the thread you can do a comparison.

Apart from me putting the sample tubes in at an angle (because of the potential condensation issue with my sample pipe shapes), the only slight difference is my chambers look a bit "taller".

Not sure what you mean by "how much heat the turbo blanket can take from the outside" - why the outside ?

But yes, keep the sample lines well away from the hydraulics, they get extremely hot.... I have bought some braided heat resistant sleeving to go on next time the engine is out (on the sample tubes only, not the chambers, they need to be able to radiate heat otherwise the sensor body temperature limit might be exceeded).

Schrödinger's cat - so which one am I ???

cheers rod ill have a good look tomorrow
with the blanket im talking about radiant heat from my drain, flanges and pipe work all of witch slightly touch the blanket
instructions say to not have any part except the white fire proof bit touch?

i need to change those fittings over tomorrow so shall see about that heat sleaveing stuff ive seen it before for insulating fuel lines tho

probaly just tap feed threads at the chambers as its thick wall think 3-3.5mm over kill but its what i had..

when the time comes to bend or in my case more like fold the feed lines ill try keep them ether up or down hill to not trap moisture as said they will most likely run down hill to the mani and the chambers already go down hill so good there

i when into my hydraulics shop and tried to exchange the brass fittings i found out that the ss ones are like 25 bucks each (need 4 plus a plug) they also said that the fittings will be good up to 1600 F? i then asked about the fittings holding onto the pipes they said bring them in if they do we will swap them for free on the spot

so my question is first where else should i look to find ss compression fittings?
second if i cant get ss whats the worst that i can happen with the brass? melt blow the line out?
not to sure what the best move is atm

The melting point of brass is about 900C so the ones in your manifold will get pretty close to that, maybe exceed it (esp. if you are wrapping it to keep the heat in).
It will go soft and deform before it melts.

Maybe they were quoting Farenheight ??? (but when I was out in Aus earlier this year it seemed more metric than the UK).

Whether the olives in the fittings manage to dig into a stainless tube (brass, copper, etc, is a lot softer than stainless) is debatable.

Re. price, I paid about GB£3 each for mine, a mixture from a mail order hydraulics company and eBay (old stock), the mixture being because I used different sizes for my EGT thermocouples.

A quick look at eBay (Aus) shows these

http://www.ebay.com.au/itm/DOUBLE-FERRULE-...=item20c4f56298
or http://www.ebay.com.au/itm/DOUBLE-FERRULE-...=item5af32dbc19

although you'd obviously be delayed by the overseas postage.

EDIT - typo(s)

Edited by Rod S on 3rd Dec, 2013.

Schrödinger's cat - so which one am I ???

will 6mm OD fitting be ok on the ss 1/4" (6.35mm OD) tubing i already have?
will that 0.35mm make it not seal or grip the tube?
im going to use 90 degree ones for space reasons mainly also less tube due to the more straight line approach un less they need to be equal length?

http://www.ebay.com.au/itm/DOUBLE-FERRULE-...=item20c4f5a2fe

Schrödinger's cat - so which one am I ???