Discrepancy between ship CHT and JPI CHT

[ArtVandelay]

Look for JPI Adapter probe. I recall two different part no. for different thread/diameter.

This what you are talking about?
https://www.jpinstruments.com/shop/m113-38-cht-gasket-probe/

[kortopates]

10 minutes ago, teejayevans said:

This what you are talking about?
https://www.jpinstruments.com/shop/m113-38-cht-gasket-probe/

No, that is a spark plug gasket probe.

see the CHT Adapter Probe listed on this page

http://www.jpitech.com/probes&sensors.php

[Rick Junkin]

49 minutes ago, kortopates said:

100f off, if accurate, is too much to be due to a spark plug CHT probe too but it could be due to the probe type setting in the JPI if it's miss-configured.

I do have an updated EDM 700 that records data, and I checked the unit programming to confirm that it is set to K-type thermocouples. Since the temps I'm seeing on the 700 are 100DF low as compared to the ship's CHT, I was hoping to find that it was set to J-type with K-type probes installed, but no such luck. It was set to K-type. I'm thinking that the Tannis probes that are installed may not be of the grounded variety, and also observed that before starting the last time I preheated the engine, every other cylinder was reading high/low. So there's a lot more going on here than just an errant probe installation. It could be that the Tannis probes have exceeded their service life and need to be replaced anyway. Fortunately its all secondary data and not safety of flight, and the observed/recorded data is consistent, so I can figure it out over the next couple of sorties and at the upcoming oil change in 4 hours when I'll have the cowl off again.

Thanks again for the point-outs, I'll report my discoveries and conclusions.

Cheers,
Rick

[carusoam]

Rick,

They should all be reading pretty steady OAT within a few degrees of each other before you start the engine.

the way they are typically calibrated is putting them in icewater reading 32°F and boiling water reading 212°F

Otherwise read up on the sensors you have vs. the ones you want...

The thermocouples are made up of two wires of different material.  The microvolt signal is generated where the two meet. No specific grounding wire for the sensor (I think).  

Grounding wires between the engine, airframe and panel are always important.  Check to see if you still have yours.

Best regards,

-a-

[Rick Junkin]

Today i took the cowl off and discovered that I have washer thermocouples installed under the Tanis heating element bayonets that feed the EDM 700. I also confirmed that the ship gauge goes to a bayonet thermocouple on #5, which also has a washer thermocouple installed under it for the JPI. The ship thermocouple has an odd knurled connector on the top of it which allows the lead to be disconnected from the thermocouple, presumably to aid in installation? At any rate, it makes me suspicious of the quality of the ship CHT indication.

I did a ground run before I took the cowl off and correlated the recorded JPI data with the ship gauge readings I recorded on video with time stamps, and show the JPI reading 285DF, 8% lower than the 310DF indicated on the ship gauge while on the ground. Assuming this is linear, I should see ~350DF on the JPI #5 when the ship gauge reads 380, but that doesn't match what I've seen before while airborne. I reviewed video and JPI data from a previous flight, and the JPI was showing 255DF when the ship gauge was showing 350DF for #5. In other words, none of the data I have makes sense because its inconsistent at similar temperatures, unless the airflow creates a different divergence profile between the two probes. That doesn't make sense to me either. The JPI CHT temperatures are very consistent across all 6 cylinders, with a spread of about 40FD across all 6.

So, more data collection. I'll also talk to JPI to get more info on the washer t-couple performance.

If I had to guess, I'd say that Rocketman probably had the same setup on his engine and changed everything out to the Tanis combined heater/t-couple to fix it, but I'm only guessing. I'll most likely do that at overhaul or at the next major engine maintenance, which ever comes first.

Or I'll figure out that the ship system is toast and just change out the ship probe.

Cheers,
Rick

[kortopates]

Rick

Are you sure they are gasket probe's under the tannis probes - do they appear to flat washer size gaskets or do they appear to be more like collar that piggy backs onto the Tannis heater. Only the latter should be used in this location and it is acceptable only if the fire sleeve is still in place at the base. But gasket probes are only used on sparkplugs - and they are terrible since they can run 30-50F hotter or cooler and you can't tell which way. But if you don't like what they're telling you, you can move them from lower to upper plugs or vice versa just so they'll read differently.   

I personally wouldn't trust what you are describing to give any meaningful CHT information. I do find  the adapter probes acceptable since they work well; providing they're installed properly according to JPI's documentation.

But its looking like your Tannis system was installed first before the engine analyzer and perhaps the previous owner at the time didn't want to spend the money to do it right. But the sensors you want for your application are the Tannis dual heater/sensor grounded K type probes.

If you don't have someone local to work on your  aircraft I can understand your desire to postpone. But this is very minor maintenance to address for the safety of knowing what your CHTs are really indicating; especially on turbo charged engine where a clogged injector can ruin a cylinder in minutes if you have no idea what's really happening with your CHTs. Its very possible your ships gauge is reading low as well.

If you do go with the Tannis dual heater/sensor probes you will also need the new style Deutch connectors to replace the old style on your older 120vac harness to plug and provide power to the newer style. Plus you will want to make sure you do have the JPI Adapter probe under your OEM CHT5 probe since that one must stay till you go to an approved for primary analyzer. That set up will give you the most accurate CHT indications. But adapter probes are acceptable if that what they really are, but not gasket probes.

Good luck

[Rick Junkin]

38 minutes ago, kortopates said:

Are you sure they are gasket probe's under the tannis probes - do they appear to flat washer size gaskets or do they appear to be more like collar that piggy backs onto the Tannis heater. Only the latter should be used in this location and it is acceptable only if the fire sleeve is still in place at the base. 

Tanis calls them washer t-couples, to be used under the Tanis heaters specifically for engine monitors - here's what's on my airplane.

http://www.tanisaircraft.com/tas2144-k.html

Cheers,
Rick

[kortopates]

6 minutes ago, Junkman said:

Tanis calls them washer t-couples, to be used under the Tanis heaters specifically for engine monitors - here's what's on my airplane.

http://www.tanisaircraft.com/tas2144-k.html

Cheers,
Rick

I have not seen these before. But a ring probe not inside the CHT bayonet hole such as this is going to be significantly affected by airflow as well as being at a distance from the intended sensor location. But given that they are $105 cheaper than the dual probe its no mystery why they sell these.

And now that I am thinking of the unique base of the Tannis heater element I am not sure the JPI piggyback adaptor probe, intended to work with the OEM probe, are compatible with the tannis heater element.  But they are accurate for their intended application.

[carusoam]

Tanis being a heater system builder for heating aircraft engines is an interesting choice...

While visiting their site, I wasn't able to navigate through their product offerings very well.

I would be more inclined to stay with JPI as a supplier even though they probably don't make the parts either.

The upside is they can guarantee that you are getting the right part for the application.

 

Of course, if you are putting Tanis heaters down the thermocouple well, and using the ring probe that goes with it...   I would understand the situation better...

My IPad was not able to get that from the Tanis site.

are you using Tanis cylinder heaters?

are you just trying to get the right TCs for your JPI?

Best regards,

-a-

 

Edited February 20, 2017 by carusoam

[Rick Junkin]

Thanks for the advice! 

The washer t-couples are also recommended in the FAQs on the JPI web site for use with a Tanis engine preheat system. I'll talk to JPI again tomorrow and see if they have any updated guidance from what's on their web page. I obviously have a discrepancy that I'd like to get resolved, and I'll take the appropriate actions to do that.

The GOOD news is that the JPI CHT indications are consistent across all six cylinders, which means I can establish a JPI CHT reference temperature by crosschecking with the ship's gauge and watch for deviation from that reference to indicate any problems. I've been using 100FD below the ship's indicated CHT up to this point. I fly in normalize mode, which simplifies that task.

Thanks again for the point-outs and recommendations.

Cheers,
Rick

[carusoam]

Rick, I just posted a couple questions above...

regarding the Tanis preheat system...

-a-

[Rick Junkin]

2 minutes ago, carusoam said:

are you using Tanis cylinder heaters?

are you just trying to get the right TCs for your JPI?

 

Yes, I have a Tanis system installed and have the washer t-couples feeding the JPI installed under the Tanis cylinder heaters that are in the thermocouple wells.

I'm trying to resolve the difference between what I'm reading on the ship CHT and the JPI CHT (JPI reads ~100FD cooler in cruise). I have a work-around for now, but would ultimately have everything indicating the same. Or at least close to the same.

I'll see what JPI says tomorrow, I can't imagine this is the first time this has come up.

Cheers,
Rick

[carusoam]

Remember to confirm the type of thermocouples you are using in each conversation with the suppliers.  As in type J or K.

Both work for the application.  JPI uses one standard and other suppliers use the other. 

The wrong type is going to give readings that will be off by a large number. Wired backwards, the number decreases as the engine warms up...

I think You may have up to three suppliers involved... a prior engine monitor, the JPI monitor, and the Tanis people.  They can all be right.  You just want them all right and on the same page at the same time for you.

don't forget to verify what thermocouple is being used for the ship's gauge...  Some industrious prior owner may have swapped it out along the way.

They All should be reading the same T when the plane has been sitting overnight...  no reason not to. OilT, CHT, EGT. Oat...

Good luck, you are almost there. 

Best regards,

-a-

[Rick Junkin]

Thanks! All the sensors indicate ambient correctly. And I confirmed K type CHT washer TCs going to the JPI. And on the ground, the JPI and ship CHTs were pretty close, as I indicated earlier. But I didn't spend any time messing with the ship CHT thermocouple, because it seems to be indicating reasonable values. I'll take a closer look when I change the oil in a couple of weeks.


Sent from my iPhone using Tapatalk

[Rick Junkin]

Ok, I've decided on a course of action. Thanks for all of the valuable suggestions.

I'm going to pull the Tanis heating elements and replace them with the "standard" JPI CHT bayonet probes. I can't do that with #5, where the OEM probe is installed, until I upgrade to the EDM 900 that's in my panel's future (more on that at another time and place), so #5 will be an outlier in my EDM 700 data. I'll have to live with that for a bit. But at least I'll have a good hack on the "true" temperatures of the other 5 cylinders.

Tanis now makes heating elements that replace a valve cover or intake manifold bolt. Pretty ingenious on their part, as that removes the conflict of which probe is going into the CHT bayonet well. I'm assuming they did it because of problems like the one I'm having, and I'm sure it will prove just as effective to heat the cylinders from those points overnight. So when winter returns next year, I'll either install the new heating element set or possibly just install a complete new Tanis system, or purchase a Quad Hornet Heater and completely remove the Tanis wiring. I'll be thinking on that one for a bit. I need to do some research on the metallurgy and heating element design of the Tanis heating element bolts to get a warm fuzzy that I won't be compromising anything by replacing a required bolt with a heating element bolt.

Thanks again for all of the thoughtful inputs and recommendations. The kicker was the repeated recommendation that I ensure the temp data I am looking at is accurate so that I can manage my engine appropriately. I have a work-around that I've been using that works, but I know how that ultimately goes. The engineer in me doesn't want to do the mental math if I can find a way to directly read accurate data.

Cheers,
Rick

P.S. I talked to JPI and for the EDM 900 TSO primary installation, you HAVE to use the JPI 5050T CHT thermocouples. The Tanis dual heater/thermocouple probes don't satisfy the requirements of the JPI 900 STC. They would work if I was going to stick with my EDM 700, but I would have to swap them again when I upgrade to the 900, and install the Tanis stud heating elements in the valve cover or intake manifold screw/bolt locations. So I'm going to incrementally change things out, starting with the new CHT thermocouples.