Fitting dual wideband 02 sensors
Turbo Mini Forums

I'm planning to put the sensor right out of the exhaust flow - I think I can lose 300 degrees with the sensor a few cm further out, so that the sensor tip is behind a small hole in the exhaust wall.

Watching the SLC Free in-car for a while should give a good enough idea of whether the temperatures are under control. I may also hook up a multimeter to the heater to one of the heater transistor legs and monitor the heater duty-cycle. This should give an idea of how close I am to the ideal sensor temperature.

Alan, any plans to output AFR and temperature (and maybe heater duty please) digitally through the unused pins? Or do you know of anyone else looking to modify the source code? I also plan to move the boards into my ECU casing. I assume there's no problem removing the displays to give me more room?

Edited by Rod S on 15th Oct, 2014.

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

I saw your question on his forum, but also saw a lack of replies to any of the handful of posts.

He's already taken the time to sign up and write a detailed post. It would be good if we can encourage his input on here from time to time and keep everything in one place. I would expect (/hope) he will check back at least a few times for responses. His current website promotion seems to suggest he's very interested in what the forums have to say. I notice he's quite active on diyefi.org, so even if we can't catch him here, then maybe that would better place to catch him.

Edited by PhilR on 15th Oct, 2014.

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

I see you're point. I think Alan himself that has the most control over his forum succeeding and I'm sure it will become good in time. His business looks young and although the support might not be there yet, he must be doing something right.

Did you use the controller in the end? Do you have any worries about it's performance?

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

I read your SLC Free thread, just couldn't tell whether you'd carried on after the problems. Interesting IOx stuff.

Because there are two heat sources, the temperature (as passed over I2C) is pinned at 750 C and only shows a change when you EGT goes over 750, and if you watch the heat output duty cycle you only know about how far below 750 your EGT is.

Both measurements are half of the same picture. For testing, having either should be much better than having no indication of temperature. Later on, I may try an idea I have to monitor both on the SLC Free, but initially, I'll just measure the heater and watch the Temperature on the screen (Shame about no I2C possibility for the SLC Free)

Has anyone tried the PLX controllers with health monitoring ?

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

Rod that information I put in the manual, there is just too much technical information about SLC Free to plaster onto my website. You got the first revision of the kit, so perhaps back then my manual was not updated with that information.

Back to the part about measuring exhaust pressure, has anyone built one and successfully tested their own backpressure monitor? If so what are the parts and fabrication procedure?

I'm just collecting ideas and parts right now. I'd hope to attempt a first version within three months.

My DVM show duty-cycle, but interesting to learn you can do it without this feature.

Roughly what temperature will the heater start to ramp down from 100%? Is it linear or PID controlled?

Edited by Rod S on 16th Oct, 2014.

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

How long until you plumb it in and get some data?

I don't want spend much time and money on this especially if you're closer to getting some results. I'm struggling to find higher rated pressure sensors for a decent price, so I'm thinking of using oil pressure senders. I've seen some that are 80psi, 1/8 npt, so I'm thinking of tapping some stainless tube (12mm OD, 8mm ID), weld that to the manifold, then screw the sensors in directly...

Oil/fuel pressure sensors often have synthetic rubber diaphragms so may not stand the heat at the distance you propose.

The Freescale MPX5700AP, same profile/pinout as the 2.5 bar sensor is rated at 7Bara so up to 6Bar in the exhaust (with the discussed heat protection).
£11.33 (+VAT) from Farnell so I'll order a couple shortly.
More for interest than persuing this quickly.
I have too many other projects on the go at the moment and although there maybe a small question mark over my sample chambers, not enough to get excited about until the engine next comes out.

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

My experience is that oil and fuel pressure sensors have very poor ability to sense low pressures, above maybe 15 PSI then they start to become accurate.

I now have a couple of the Freescale 7bar(a) sensors sitting next to me (Farnell are quick), 4mm x 1/8 fittings ordered (eBay China so about 2-3 weeks but they were way cheaper than my usual UK source) and I'm waiting to hear back from my UK source whether they have 4mm x 0.5 wall tube, currently they are only showing 4mm x 1mm wall which might be a bit restrictive.
And I have plenty of 4mm silicone tube.
And obviously I have all the electronics to do the measurements.

This will be a tight squeeze to fit - I currently still only have the original Metro turbo cast exhaust manifold so space will be tight - although that will obviously be good as it's the worst case scenario for pressure, but this won't happen until the engine is back out which won't be until the back end of this year, really I'm just collecting parts too to save time later.

Just to be clear though, all I will be doing is measuring the pressure. I will also re-commission the EGT thermocouples (assuming the damaged cable one still works) which I haven't bothered with ever since Paul and I found no correlation between EGT and AFR, but there is simply no room to put 4.9 LSUs into the Metro manifold (I do actually have two now) and, even if I could, my SLC-OEMs only work with 4.2s and Alan says it will be a while before there is a 4.9 version.

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

Some further thoughts re. temperatures up a pressure sampling tube.

I think Phil and Paul have show it's sod all in a static environment but I'm still not convinced in a dynamic environment.

The exhaust gas is compressible so, although the totallity of the tube is a dead end at the sensor, if there are pulses in the exhaust gas, the tube will see flow backwards and forwards, greatest at the manifold end tapering to nothing at the sensor. That continually reversing flow could "carry" the heat further up the tube.

I have found a source of both 4mm OD x 1mm wall and 4mm x 0.5mm wall. Thermal conduction will be less in the thinner wall (assuming the heat is being disipated to the surrounding air) but I'm also thinking that the pulses could transmit heat further up the larger bore thus cancelling out the benefit of the thinner wall.

It may come down to practicalities, I suspect the thinner stuff will kink when trying to form it into a coil (even with a proper pipe bender) and I can stick a couple of thermistors on either and measure the temperature (if the thin stuff will bend OK) but any thoughts in advance ?

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

I could do a transient simulation, but some of us are actually working to get our cars on the road

The results would be fairly predictable though.

If you are running 1 barg boost, then the average pressure in the exhaust manifold could be as high as 3 barg, varying by 1 bar each side of that.

So with a pressure range of 2 barg to 4 barg, then the air will compress to around half it's volume through the cycle.

Just make the tube twice as long as you originally thought.

Now, where that brake master cylinder I was looking for

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."

Picking up this thread again, I got my 4mm x1 and 4mm x 0.5 and surprisingly the thin walled stuff didn't kink



I have yet to figure out how to get the coil beyond 270 degrees without machining up my own former...

Then, today the fittings arrived from China



It looks like 4mm fittings (with an 1/8" male thread) should be viable and compatible with all previous 6mm stuff.

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

I've had a go at this over the last couple of days too. I started with a pipe bender like yours but also found it limiting. I then tried bending it around any old former. This was bent around a socket extension (about 1/2" diameter), pipe is 3.3mm , 0.5mm wall; no signs of kinking or crushing:



I was trying to do away with the compression fittings to save room as I need space for the o2 sensors too. As per the photo I wanted to make an interference fit in a sensor boss (3.3mm tube in 2.25mm hole) but I'm not happy with it. Your compression fittings look like a better idea, so will likely get some of them, and 4mm tube.

The photo show the post turbo boss, but for the pre turbo ones I'm looking at 'De-cat sensor extenders' which are longer and some have restrictor holes. I've also seen that AEM sell medium length bosses for pre turbo installation; they appear to have groves cut into the body, to create cooling fins.

http://www.aemelectronics.com/products/sen...-o2-sensor-bung

With regards to the pressure sensors, Rod, I think I'll get the same ones as you. In my head, I thought they were more expensive with a lower pressure rating, but from what you and Jean have written, this seem like the best option

Edited by PhilR on 25th Oct, 2014.