Cylinder Head Temperature Gauge - Should I? & or Air Fuel Ratio

Mark, that does not apply to the kind of thermocouples we use which we cannot calibrate. If you extend the length of the provided thermocouple using thermocouple wire you will mess it up as the gauges are calibrated to use specific length thermocouples, usually about 2 feet long. You must extend with ordinary copper wire to maintain the calibration.

 

Yep, it's amazing they work at all considering how far away they are. Yours was an aircraft one wasn't it Zedders, or am I getting conf..?
It was rather snazzy from what I recall.

 

I believe the thermocouple that's matched with my Westach gauge is a J type. Indeed looking at the Westach website shows their current range of gauges intended for light aircraft use use J type thermocouples. Any idea why they would choose that in preference to K or T types? I did look into this briefly when I got my gauge but didn't get far before baffling myself!

 

You don't understand how they work if you think that. What you use at work is completely different because you can calibrate it + there is an accuracy advantage to having longer wires
A thermocouple is simply two pieces of wire with different resistances. The gauge reads the difference between the two.
We know that the longer you make a wire, the more resistance it has - that's how your tank sender works.
If you make your thermocouples longer by extending them with a random length of thermocouple wire you increase the resistance of both wires according to their type so increasing the difference between the two that your pre-calibrated gauge will not like at all. Trust me - I tried it, the gauge barely twitched n a cup of tea. Not surprising as I was giving it about 8x the difference in resistances it was calibrated for.

For you in your laboratory, extending with thermocouple wire in effect extends the length of the thermocouple providing larger swings in measured difference - good for accuracy. It also allows you to site the cold junction in controlled conditions. For us plebs it's a gauge pre-calibrated to a 2 foot J-type thermocouple which is deemed accurate enough.

 

*it's about 2ft - other gauges may vary slightly.

I totally get that, but it's still the wrong advice for car gauges.

 

If you look on the Westach website and read the labels on the thermocouple pictures, they are all J-type 48" long, they just have different fittings for different circumstances. I guess you probably need a bit more accuracy and longer cables on an aeroplane.
https://www.westach.com/cht-thermocouples

 

The CHT gauge is early warning.. you have a lot of extra power. Which is producing extra heat. I would go CHT. Your damage was perhaps getting the heads so hot the aluminium started to soften (150C and up !) letting the valve seat loose. The guides should not have been loose. I would have a word with Roy at Mofoco (hes a regular on the Samba, as its unusual for the guides all to be loose.

Your CB mixture gauge would be using an old narrowband sensor. About all thats good for is setting idle mix in the middle of the dial..neither rich nor lean. The rest of the time the mixture would be rich except for cruising ..

Of course these days a 5 wire wideband O2 sensor only costs £25 from China.. i replaced the Bosch sensor on my PLX air fuel ratio meter that faded out at 50k miles with this. Wrong connector but easy to bodge.

 

@paulcalf This is what you want, with 4 thermocouples just $500.

 

Thanks, I had seen that and think mine came with a 5WK lead. I just had to make a new 14mm copper ring for it - carefully crimped, not soldered on! Handy having 48" to play with, made it easy to end up with the cold junction somewhere with a reasonably stable temperature (spare wheel well, under my water tank). As for accuracy, well all I could really do was test it in boiling water where I did see bang on 100*c.

When driving it's amazing how much and how quickly the CHT responds to engine load. Gentle cruising (55mph on the flat low 20s ambient) sees about 150*c, give it some hard work and within 10 seconds it's up around 170*c. Gently slow down for a junction or roundabout and within seconds it drops below 140*c, bumble down a steep hill and it's down to 125*c. Fascinating to observe and can almost become a distraction but the novelty hasn't wore off yet .

My oil temps generally hover around 80-90*c once warmed up and driving normally (VDO dipstick gauge) with any visible variation taking maybe 5 minutes to show up.

All the above with stock 1600 dp, Mexican heads, std carb (130 main) and quiet pack exhaust.

 

The thermocouple cable (K Type) on my sensors has pretty serious insulation (fiberglass braiding with stainless steel overbraiding). I’ve been assuming that is due to the high temperatures at the source e.g Cylinder Head and Exhaust.

I make too many wrong assumptions, so a couple of questions:
•The cables don’t actually carry a temperature, just a ‘signal’, so can they be safely put inside a conduit alongside normal 12v cables to the gauges?
•Apart from by the sensors (which i’m not doing anything with), can I use normal 12v heat shrink tubing to tidy up connections that i’ll be making?

Thanks.


Sent from my iPhone using Tapatalk

 

Nooooo .. the metal thermocouples and their wires is made of is resistance wire.
If you have a long wire and a microAMMETER on the far end (no power just thermocouple current moves needle) then extending the wire makes it read lower.

This is traditional as its a meter and some wire with a second thermocouple at the meter. No electronics just works.

As the meter movement is mechanical it averages out interference naturally so you can run wires to these gauges alongside other wiring in the bus.

Copper is lower resistance than thermocouple wire so it has less voltage drop. BUT every join metal to metal in the circuit makes another thermocouple producing a voltage. So you have to make all connections identical and with both sides at the same temperature ( wrapped in insulating tape bundle e.g.)

If you have a VOLTMETER on the far end then practically no current flows so the length of the wire doesnt matter. In that case the gauge will be powered and contain more electronics.

I use MAX5575 Arduino modules that have a voltage sensing chip so the thermocouple wire lengths to the chip dont matter. But because it samples values periodically, it cannot average interference like the simpler microammeter setups so it gets interfered with by ignition noise..
However I can extend the wires keeping accuracy as the wire to the dash carries digital value representing the temperature not thermocouple voltages or currents.

 

This was the chepest Westach i could find when looking if anyone is searching..

http://www.skydrive.co.uk/products.asp?cat=59

Need to sort mine needles still not moving



Sent from my SM-T580 using Tapatalk

 

I think reading through all the comments above answer your question @paulcalf , keeping things simple with just VDO oil temperature and pressure guages is the way forwards...

Can-o-worms

Sent from my SM-G960F using Tapatalk