Monitoring all 4 CHT/EGT

[fastj22]

This is what I've been thinking of doing on my aerovee, any reason it would not work?

I'm going to start my wiring soon, and have the switches already,

You might need to use relays. Each thermocouple has its own return (ground) to reference the resistance. So for each thermocouple you would need two dual throw switches. One common from the instrument tied to each lead. The cockpit switch would energized the relays of both switches which would connect both leads from the thermocouple to the instrument. Normally open, it would read one sensor, when energized, it would the other sensor. Repeat for as many as you want to read.

[Bryan Cotton]

The problem with using switches with thermocouples is that every time you make a junction with a dissimilar metal (like copper) you get a voltage change that messes with your temperature. If you get a proper TC switch box you are looking at $175 for 6 channels or so. We used to use the Pico Tech TC-08, a neat little box but runs about $400 for 8 channels. If you are clever you can find an Analog Devices chip that will condition TCs and you could read it in with a PIC microchip or something. That is what I was going to do with my hummelbird. RTDs are a different animal, they are 3 wire copper conductors and you can switch them.

[gammaxy]

I suspect a mechanical switch can work in this application if the switch is close to and at approximately the same temperature as the measurement instrument (as is probably the case when both are on the instrument panel).

If assuming the switch, instrument, and copper wires between them are at a constant temperature is too big of an assumption, short thermocouple extension wire could probably be used from the switch to the instrument and the "Law of Dissimilar Metals" invoked on the simpler assumption that the switch is at a constant temperature inside.

It would be wise to test the arrangement out at known temperatures and even try to intentionally induce an error by heating the switch independently of the instrument.

[Bruce593SX]

Could you wire some sort of switch (DPDT) to monitor all four? The switch would basically choose which thermocouples are connected to the instrument.

One way it monitors front cylinders, switch it the other way and it monitors the back ones.

I emailed MGL support and received the following response:

Matt Liknaitzky (MGL Avionics)
Jan 07 06:48 PM

Hi Bruce,

This can be done, as long as the probes are the same type.

Best

Matt

MGL Avionics US
3401 Airport Dr
Torrance, CA 90505

So I'm gonna install them on my stratomaster E1 switching between front/back. Won't have first flight till sometime this spring. Will report back

[Sonex1542]

Let me see if I can add more confusion to this. I've spent the last 6 months looking at my own EFIS design. Trying to roll my own here! Should have the capacitive fuel probe complete in a couple weeks to feed it. Parts on order $20.00 Princeton Probe :D

MGL & Dynon at least are using a not so simple RDAC (read: custom) that has both A2D and separate MAX31855 (most likely) chips to process the thermocouples. Best way to handle adding more thermocouples than channels is to switch it via a relay close to the RDAC.

8 channel MAX31855 is not cheap. But for about 100 bucks can chain 2-4 channel spi to raspberry pi and go that route. Nice small display.

In any case... Use a relay (mechanical switch), keep lengths short. Or roll your own.

[Sonex1542]

You might need to use relays. Each thermocouple has its own return (ground) to reference the resistance. So for each thermocouple you would need two dual throw switches. One common from the instrument tied to each lead. The cockpit switch would energized the relays of both switches which would connect both leads from the thermocouple to the instrument. Normally open, it would read one sensor, when energized, it would the other sensor. Repeat for as many as you want to read.[/quote]

Type K thermocouples don't measure resistance, the actually generate a very small voltage. MGL has a nice RDAC they've put together. I think it's listed for $300 and change without the head unit. I'd love to get my hands on the data sheet for it to be able to read from it. Programming would be a piece of cake!

[gammaxy]

I'd love to get my hands on the data sheet for it to be able to read from it. Programming would be a piece of cake!

Here you go: http://www.mglavionics.co.za/Docs/RDAC% ... otocol.pdf

MGL is pretty open with their protocols, but sometimes the documentation is a little out of date.