john banks

Some suggest drilling the turbo housing which I am not willing to do/have done. Others suggest the headers - outlet #3. The question is how/who to do it? Apparently TIG welds melt with EGT type temperatures of up to 900 C. Others drill a hole and use a tap which they then weld on.

What do you recommend? These are some dimensions of an RS device - do you shove the whole 25cm into the pipe?? Do you bend it at the bulb end?



[Edited by john banks - 4/28/2002 3:30:20 PM]

BugEyed

John

Don't use that type of probe - it will fail and will go through your turbo. Whilst damage is unlikely it may occur, but at very least you will have to replace the probe on a regular basis. This design is too small a diameter for the unsupported length in the gas stream, and will suffer from flexing in the exhaust gas pulses (not a nice environment - 800C and gas flow pulses at 50 Hz +).

FWIW, you mount the probe so that the tip is approx 35% across the diameter of the pipe to get an "average" reading. The best type of mounting is one that you can remove to change the probe easily. Hence, people either weld or screw a compression fitting to the exhaust. Being a belt and braces type of chap, I'd suggest you drill and tap the manifold, screw the fitting in, and then weld the fitting to the manifold. This way the weld is only to stop the fitting unscrewing itself.

I suggest that you save your pennies up and invest in a "proper" EGT probe.

Duncan

[Edited by BugEyed - 4/28/2002 11:46:29 AM]

john banks

Thanks, proper EGT probe - where, what make etc?

mutant_matt

John,

This probably won't be very helpful but EGT gauges usually come with a suitable sender and mount. I know Defi's do but not sure about others.

What were you planning on using to do the monitoring? (you know you want a set of Defi's )

Matt

john banks

Maybe I'll speak to Jamie I was thinking of RS52 gauge from RS which takes input from standard K thermocouple plugs and reads max, hold, etc and will do my intake temp as well. They are only temporary whilst tuning. Cosie Convert is inspiring me with his tales of nice cool EGTs at AFR's I would not dream of

[Edited by john banks - 4/28/2002 12:50:03 PM]

dowser

I'd heard that the up-pipe available from Scoobysport for >my01's will also fit <01's - it removes restrictions, and has a nice lambda boss built in....ready for an egt probe?

Pete C. - can you confirm?

Richard

R19KET

John,

I've used the RS K sensors for several years, and they are the same probes supplied by SpA with their EGT gauges.

However, I would recommend using a min dia' of 3mm, based on Duncans comments, and my own experiences, but I would only tend to use about 3/8 ~ 1/2". You can bend them, but not a right angles.

RS also do the compression fttings for them 381-734 (3mm) or 381-7356 (6mm).

I did have problems with the 1.5mm sensors loosing the tips, and spoke to the manufactures. They said the main problem was the "heat cycling", and recommended the 3mm, which seem to have worked.

Mark.

john banks

Thanks! What about using a BSPT version of the same so it has a tapered thread? Does it need to be welded as well?

Need to negotiate RS search engine to find a thermocouple - 3mm diameter - presumably just long enought to got 1/3 - 1/2 way into the pipe? What sort of diameter is the manifold?

Looking in RS under "All Products | Automation | Sensors/Transducers | Temperature Sensing | Thermocouples" I cannot seem to find what I am looking for - anyone help? www.rswww.com - the full URL is too long to paste into Scoobynet and work!

I am using the exposed junction K with fibreglass for intake and that plugs into RS 51 or 52 meters very nicely.

[Edited by john banks - 4/28/2002 3:16:24 PM]

john banks

I've found 397-1264 but the probe is 3 x 150mm



Mineral insulated thermocouples able to withstand temperatures up to 1100°C (pot seal max temp 180°C). The probes can be bent to enable fitting in awkward locations and are particularly suitable for industrial applications including high pressure, high vacuum and high vibration. The thermocouple junction is located at the tip of the sensor and is insulated from the sheath.


Type K mineral insulated thermocouples
Stainless steel (310)
Very flexible
Plain pot seal
1000mm PTFE flat pair 7/0.2mm cable
Max temperature of probe 1100°C

Is this suitable? 15 cm probe - how do you bend it?

R19KET

John,

The fittings I gave the pt.no's for are parallel.

Personally. I cut the thread off, weld the fitting on, and then drill through the fitting.

Mark.

David_Wallis

Theres no problem welding to the exhaust... as the exhaust has weld's on it allready... I would just drill a hole and then weld a nut over... and just mount like a lambda sensor...

I certainly wouldnt try welding the sensor.

David

john banks

Thanks. Is 15cm too long for the probe or do you just poke say 4 cm through and bend the rest?? Sorry if I am being thick, just want to know exactly what I am doing - I'll get someone to do the welding for me though, but I need to get the right bits.

R19KET

John,

I only poke about 1/2" through, and then bend the rest, just don't bent it too sharply. The bends need to be radiused (sp)

David, you don't weld the sensor, just the fitting.

Mark.

john banks

OK, so the other 13.5 cm you just tie down having bent it smoothly?

[Edited by john banks - 4/28/2002 5:52:24 PM]

David_Wallis

ahh now I see..

Still you arent going to melt the welds with an exhaust..

David

john banks

Welding to you is like circuit building to me I am sure. I wouldn't dare, doing a tap will challenge my skills

David_Wallis

Youve seen my thread then? Ive got the best of both worlds I can build circuits given a diagram or instructions... just cant design anything complex... But on the other hand my dads background is electrical and mechanical engineering.... he's an electrical design consultant for an electricity board... upto 132kv... ouch!

David

BugEyed

John / Mark

Using a 3mm probe is a compromise. If Mark says that they last for him at 3/8" to 1/2" I'm sure he is correct, but they will not be reading the correct EGTs. The flow at the edge of the pipe is slower and cooler than that at the centre of the pipe, pareticularly at lower revs and part throttle (=lower gas speeds). If you want absolute accuracy then you should put the junction of the probe approx 35% of the diameter accross the pipe. On the uppipe in my MY01 that equates to 7/8". The problem with this is that you are leaving both a much greater area exposed to the heat and force of the gas flow, hence the force on the point where the probe is gripped is greater. A 3mm probe might fail in these conditions. However, despite saying all of this you will have a reasonable measurement at the length used my Mark, and as long as your EGT target is not too high then you will be OK.

In terms of mounting the probe, as long as you bend the probe around a sensible radius (2" +) then you'll be OK. Dowsers suggestion of using the MY01 uppipe replacement is sensible provided you can get an olive to mount your probe in the standard (to an MY01) mount. I would also caution you to consider D T-S's comments about uppipes on MY02 STis, which to an extent apply on all of the cars when replacing the uppipe (although mine is currently OK - touch wood).

Hope that helps (and doesn't hinder too much).

Duncan

R19KET

Duncan,

Fair comments. I think the main issue, is finding a sensor that will give a reasonably accurate idea of egt's. Nobody should be tuning the egt to the "enth" degree.

I agree about sensor position, but then we can start getting into position in the system, and turbulence, etc'.

I would always suggest that leave a minimum safety margin of 75C degrees. People also need to know just how much the temps vary, depending on where the sensor is fitted.

I used to have 4 sensors fitted in my exhaust, from the collectors, to the D/P, and saw differences of up to 300deg C over that distance.

Mark.

john banks

It is the relative values that interest me in particular. Maybe 3mm is the best compromise for temporary monitoring.

Jay m A

iTrader: (2)

Great thread.

Especially as I am in the process of deciding where to fit/how to mount the EGT probe.

I will probably get a phantom gauge from Scoobymania and fit the probe provided. I assume the probe will have certain thread size (is it a common size? and if so, does anyone know what it is?). My plan is to move my Lambda sensor from the exhaust manifold (MY95 WRX) to the downpipe, then fit the probe in its place. I can get an adapter made so this will fit, and if theres enough interest I'll make up a batch.

It is interesting to read the differences in EGT depending on where the probe is mounted, however like John I'll use it for relative measurements only.

Justin

RICH WILD

Hi,

I've just had a HKS Electronic EGT fitted. Gorgeous bit of kit, but I've had the probe fitted in the downpipe a few inches behind the lamdba sensor. Is this too far back in the system as I've noticed that in normal driving the gauge doesn't move from its 500 degree starting point and onlt when you really go mad does it go up.

Got it as high as 725 degrees this weekend. Is that dangerous at a point so far away from the engine?

Would I be better resiting it in my new HKS manifold (on its way)?

I was just worried about what Mark said about if anything breaks off and flies into my turbo

Cheers

Rich

BugEyed

John / Mark

The position of the probe is important if you are tuning to the maximum degree, but it also depends upon what your opinion of a safe limit is. On race engines (the whizz bang qualifying sort) I've happily run 1100 C - and they did exactly what they said on the tin, serious thermal damage after 4 - 5 laps but + 50 HP over normal race settings of 950C. For a "long life" road engine I wouldn't like to see over 850 - 870C measured close to the exhaust ports with the probe in the optimum position. You might get away with more, but I'd be worried.

John

If you are doing something temporary, make sure it is easy to get at to remove. I suggest that you drill the exhaust headers (there is a nice boss by No3 I notice on my MY01 - is it there on your MY00?). If you haven't already had them off to port them, why not do both jobs at once. If you have had them off (oh, err), well you now know how easy if is to do it.

Rich

Measuring the EGT in the downpipe is less accurate - you will suffer both a big (but unquatifiable - although Mark noted 300C in an earlier post) temperature drop and a significant lag. Therefore any problems will be hard to quantify and it may be too late by the time you have!

Still, if you are getting a set of HKS headers, take the opportunity to move the probe now.

Duncan

john banks

Have ordered the following - Cosie I think you said you had a bit of spare K wire to extend it - can you just solder join K thermocouple wire? I got a tapered compression gland which might just hold nicely with a quick weld I hope.

397-1264 1 Sensor, thermocouple, K, mineral insulated, 3.0mm dia, 150mm length, 1000mm lead £15.45 £15.45
158-610 1 Gland, compression, stainless steel, 1/8 BSP, 3.0mm dia bore £8.53 £8.53
219-4955 1 Connector, plug, std, line, K £2.95 £2.95

BugEyed

John

So where are you going to mount the compression fitting?

Duncan

john banks

I'll look on the outlet to number 3 if we can get the thing off - will be in powerful spanner mode after removing the downpipe and turbo, and I will be wearing dirty clothes which always helps loosen bad bolts

Cosie Convert

John

Suggest you buy the proper plug and socket to join the wire, they are only pennies each OK maybe a pound but still in your budget I have plenty spare cable.

I agree it is relative temp you are watching, start off with conservative boost levels and monitor as increasing......but then you know that.

Andy

john banks

Didn't think of that - and order now through. I'll wrap the ends together and use solder just to seal them in position so hopefully there will be no funny effects. Weird things thermocouples....

Cosie Convert

Rich

I'd be a bit wary of 725 deg C in the downpipe

RICH WILD

CC,

What's a safe temp to run in the downpipe, just out of interest?

Gonna move the probe into the headers when I swap them, what's a safe temp there?

Cheers

Rich