Discrepancy between ship CHT and JPI CHT

[rocketman]

I have noticed that  all 6 of my JPI CHT probes on my Bravo runs very low low around 230 to 260 while my ship CHT probe runs typically 375. Which probes are more accurate?  Do they have separate k probes for different applications?  The JPI probes are attached to the cylinder directly and not around the spark plug which I think is a better way to measure the temps. Any help is appreciated. 

Ron

[takair]

Just based on the statistics, one might guess that the ships probe is the one that is wrong.  However, one can't assume.  There is typically a difference between cylinder mount and plug mount, but this seems like a large split.  First thing to check is to be sure you don't have an exhaust leak onto the ship probe.  This can drive the number up.  Second would be to calibrate it.  You could simply have them tested in boiling water or hot oil....just do the factory probe and the JPI probe on that cylinder...

[Guest]

Bayonet probes are the most accurate method of measuring CHT, all others are less accurate.  The factory value sounds like a more reasonable value, I have a hard time believing the JPI numbers would be so low.  

Clarence

[co2bruce]

When we turn on our master our temps all come up. The JPI all read the ambient temp exactly, and they all match. The ship's CHT reads about 10 degrees higher.

[mike_elliott]

50 minutes ago, M20Doc said:

Bayonet probes are the most accurate method of measuring CHT, all others are less accurate.  The factory value sounds like a more reasonable value, I have a hard time believing the JPI numbers would be so low.  

Clarence

I spoke to Ron yesterday about this, it might make sense if the wrong probes are being used for the JPI, but that is about all I could hypothesize would lead to this anomaly

[carusoam]

Note about probes:

1) They are made of two different wires, two different chemical substances.

2) They generate a microvoltage (thank you Dr. Bill) where the two different materials contact each other.

3) the color of the insulation is a standard that indicates which type of wire is contained inside.

4) the two most common TCs are J and K.  J type is typically lower in material cost and has a different temperature range than K type.  

5) the TC's wires are often an expensive material, it is mostly the fancy spring loaded mounting device and the person that installs it that add significantly to the cost.

6) One big variation is the location where the junction of the two wires is.  They are twisted together and soldered (welded?) to immobilize them. Mounted inside a sheath, then installed on the engine using the pressure of a spring or screwed down mounted device like the spark plug or other TC device.

7) A huge variation is where the TC is mounted on the cylinder.  Temperature of the cylinder varies all over the place. Hottest near the exhaust valves and cooling toward the furthest fins.  

8) Smaller variations occur with cleanliness of the installed device.

9) errors happen when owners use the wrong TC types, install the wires backwards, or improperly extend a wire...

10) checking the calibration is as simple as putting the probes in ice water and boiling water.

11) Some adjustments may be made from the JPI to make the probes read a similar value...

12) After the engine has been sitting for a day or two in the hangar.  The temps will be as close as possible. Usually within a degree or two. Include what the OAT probe is saying in your comparisons. Is your OAT being displayed by the JPI as well or a completely different device?

13) a ring probe under a spark plug often reads 50°F different than the bayonet probe in the CHT hole.  This is more an artifact of the probe's location than it is the probe type.  A similar ring probe is available to fit under the bayonet's holder.  

14) the ship's CHT gauge has one documented legal location.  This leaves the JPI's fourth probe no place to go. A ring probe is an old standard solution that worked pretty well when fit under a spark plug. A ring probe that fits under the hardware of the CHT hole is placed in an even better location.  A bayonet with a dual TC in it would be best.

Just some PP thoughts towards TC challenges. I am not a mechanic...

Best regards,

-a-

[Hank]

31 minutes ago, carusoam said:

Note about probes:

1) They are made of two different wires, two different chemical substances.

2) They generate a microvoltage (thank you Dr. Bill) where the two different materials contact each other.

3) the color of the insulation is a standard that indicates which type of wire is contained inside.

4) the two most common TCs are J and K.  J type is typically lower in material cost and has a different temperature range than K type.  

5) the TC's wires are often an expensive material, it the fancy spring loaded mounting device and the person that installs it add significantly to the cost.

6) One big variation is the location where the junction of the two wires is.  They are twisted together and soldered to immobilize them. Mounted inside a sheath, then installed on the engine using the pressure of a spring or screwed down mounted device like the spark plug or other TC device.

7) A huge variation is where the TC is mounted on the cylinder.  Temperature of the cylinder varies all over the place. Hottest near the exhaust valves and cooling toward the furthest fins.  

8) Smaller variations occur with cleanliness of the installed device.

9) errors happen when owners use the wrong TC types, install the wires backwards, or improperly extend a wire...

9) checking the calibration is as simple as putting the probes in ice water and boiling water.

10) Some adjustments may be made from the JPI to make the probes read a similar value...

Just some PP thoughts towards TC challenges. I am not a mechanic...

Best regards,

-a-

 

"I am not a mechanic..." but you sure know your industrial equipment! Very succinct account of thermocouples, Anthony. I'd expect JPI use J-type thermocouples myself. The manuals somewhere should say this. I've seen equipment set up for J get used with K, and it just doesn't work right. This is something else to doublecheck.

[MyNameIsNobody]

Moonfusious say: "Probe and an answer you shall find"...

[1524J]

4 hours ago, rocketman said:

I have noticed that  all 6 of my JPI CHT probes on my Bravo runs very low low around 230 to 260 while my ship CHT probe runs typically 375. Which probes are more accurate?  Do they have separate k probes for different applications?  The JPI probes are attached to the cylinder directly and not around the spark plug which I think is a better way to measure the temps. Any help is appreciated. 

Ron

Personal experience.....we installed a insight engine monitor and attached the spark plug probe (ring type) to the  insight monitor while routing  the bayonet probe to the analog gauge. This was on the #3 cyclinder. It would read 40-50 degrees higher than the other CHT's (bayonet probes) on the insight monitor. After switching them, all CHT's are within 10-15 degrees of one another. For whatever reason spark plug probes will give higher temperature readings than bayonet probes. 

[carusoam]

I added 11), 12), 13) and 14) to cover the ring probe and other issues...

JPI lists K-type TCs...  GEM / Insight lists J-type...

Bayonet: https://www.jpinstruments.com/shop/cht-bayonet-probe/

(Bayonet location) Ring: 3/8" (?) ring http://www.aircraftspruce.com/pages/in/probesandsenders_cht/1105561.php

Spark plug Ring: 5/8" (?)  https://www.jpinstruments.com/shop/cht-gasket-probe/

Details regarding actual wire chemistry/types...

JPI uses K-type thermocouple wire. ANSI* color code of Red and Yellow in a Yellow jacket:
‘Red’ (-) (Nickel-Aluminum or just “Alumel”)
‘Yellow’ (+) (Nickel-Chromium or just “Chromel”).

GEM/INSIGHT uses J-type –ANSI* color code of White and Red in a Black jacket: ‘White’ -Iron –positive (+) magnetic
‘Red’ – Copper-Nickel – negative (-) “Constantan” –non-magnetic.

Consider these as ideas from a PP.  check with your mechanic for suitability....

Best regards,

-a-

[carusoam]

Comparing to IO550 experience...

I get CHTs in the 320°F range in summer...  280°F in the winter.  One cylinder #5  is different, it is hiding behind the alternator and air flow is different because of that.

To really get down to 230°F, I would be running low power LOP with really cold OATs...

Keep in mind these are from old fuzzy memories, too...

Best regards,

-a-

[kortopates]

Suggest you Review your JPI configuration parameters in program mode. There is a setting for which type of probe you have installed. This allows the JPI to work with other types of probes (I.e. Both J and K type). So you need to be sure your configuration is set properly for the type of probes you have installed - otherwise the indicated temp can be a 100F off.

Regarding spark plug gasket probes - they can read either 30-50F High or Low. We see them both ways. Because of that they are really worthless and should be replaced by an adapter probe that piggybacks onto the OEM CHT probe to provide an accurate temperature indication.

Sent from my iPhone using Tapatalk

[Guest]

6 hours ago, MyNameIsNobody said:

Moonfusious say: "Probe and an answer you shall find"...

Isn't that a line for a proctologist?

Clarence

[carusoam]

Paul,

I was unable to find a JPI supplied 3/8" adapter probe.  The one I found was from another aviation supplier at Aircraft Spruce. Any idea if I have missed something?

The 5/8" ring for the spark plug that JPI has, mentions a temp difference between top and bottom plugs as well.

Best regards,

-a-

[larryb]

I flew my plane about 400 hours with a JPI 700 before upgrading to a JPI 900 this year. The 700 install had a piggyback probe on the factory CHT probe for #3. The other 3 probes were the standard bayonet CHT probes. The 700 always read #3 as coolest. With the 900 install I have the same bayonet probe in each cylinder. Now the #3 reads the hottest. I think the difference is around 40 to 50 degrees F. Note that I'm not talking about the sparkplug piggyback which is well-known to read hot. In my case the CHT piggyback read cool. 

Now, back to the OP. In Rocketman's case, we are off by over 100 degrees F. That is a lot regardless of probe location. I would first check the temps prior to start and compare to a trusted temperature reading of the engine. After that I would take one of the JPI probes, bind it to a trusted thermometer probe, and check it at a higher temperature such as a pot of boiling water. It should be easy to prove which gauge is reading the proper temperature, and go from there.

Larry

[tls pilot]

many points have been made and in my expercience is correct.

The ring adapter under the spark plug is usually what makes the ships CHT reading.

The 6 probes most likely go to the JPI, it could be possible but odd that the ring adapter and 5 probes go

to the JPI.

The other critical point is to make certain that the JPI unit is J or K. It can differ. In my case, I had a GEM/Insight

when I traded up to the JPI rather then change all probes, Joe( The JP in JPI) converted it.

A year ago I sent it back in after years of use for a new display, the question came up again.

Personally I have great success with Alcor probes - now owned by Tempest.

Lastly, a good test to start with is after it sits in your hangar overnight, turn on the master and record the readings of each 6

cylinders on JPI and analog ship guage ( may be difficult due to markings on ships guage)

[FlyDave]

Ron,

What power settings and fuel flows are producing those temps?

Dave

[mike_elliott]

Ron, another thing to check is the Mechanic's connection of the probes on the JPI. I assume he re-used the existing probes from your original engine, and I didn't notice the JPI being off when we flew together a while back. Did he put the star washer between the 2 eye connectors of the probe and the harness? While I still stand with my hypothesis it is the wrong probe or the wrong JPI config of probes, this should be checked also.

[Browncbr1]

I have collar probes that go on with my tanis heater probe, which is the same type that goes on the factory CHT... There is a copper smash ring that pushes the collar further against the cylinder..  I found seemingly hotter temp readings after tightening the probes down.  I think the person who installed my JPI 830 didn't use the 1/2" slotted socket, therefore didn't torque them down enough...  After that, I also reinstalled fire sleeve and go them zip tied very tightly... Seems more accurate now..  especially on #3, where the factory CHT wasn't very tight.

I've wrestled with if my choice to have the JPI 830 and keep redundant factory gauges was the best choice over the 900...  Recently, I noticed high MAP on the ground, crossed referenced JPI and factory the same... it was nice to figure it wasn't just the sensor.. I found a broken vacuum flair fitting. 

[Rick Junkin]

Can anyone tell me how this was resolved? JPI EDM 700 CHTs consistently 100 degrees less than ship CHT gauge?

I have the exact same issue with the EDM 700 in my TLS and can't find a way to resolve it short of replacing all of the CHT bayonets and hoping that works. I'm not a fan of "HOPE" (High Output Performance Expectations) as a strategy....

Cheers,
Rick

[carusoam]

+1 on identifying where all your sensors are located.  

The cylinders all have one probe well and the world has refused to put two thermocouples in the one well.

The ships gauge rightfully has one well reserved for it.

Expect that a previous owner may have improperly set the JPI up to show uniform temperatures on the digital screen.

Proper operations and limitations require the Ship's CHT to remain in the prescribed location.

An improved thermocouple for this challenge is a ring probe that gets mounted under the well's TC mount. Piggyback style.  This location is usually about 25°F hotter than being in the well.

A really common misguided approach to this challenge gets a ring thermocouple that doubles as a plug seal. This is great for cylinders without TC wells.  This location is about 50°F hotter than the well, and is susceptible to damage each time the plug gets removed...

TCs are laws of physics followers they can't be 100°F off.  But the indicator on the panel has a calibration option that may get adjusted...

Sound familiar?

With a 100°F difference, expect that there is a rig probe sitting under a spark plug somewhere on the engine.  Twelve locations to look...  

While you are in there... Give a good look at the exhaust tubes and clamps that supply the Turbo. These have some maintenance procedures that are good to understand as an owner.

Best regards,

-a-

Edited February 17, 2017 by carusoam

[Rick Junkin]

I need to drop the cowling and have a look to see what's what, but in my research I believe I may have uncovered another issue with my particular installation that I need to consider - I have a Tannis engine heating system installed. I'll confirm it when I inspect the engine, but my understanding is that the Tannis system uses the CHT bayonet positions for the Tannis combination cylinder heating/CHT thermocouple installs. They are supposed to be K-types thermocouples, but my airplane is a 1989 and I haven't reviewed the log books to see when the Tannis system and the EDM 700 were installed and what exactly is in the thermocouple wells. I have a sinking feeling that I'm looking at about 1.2 AMUs in CHT thermocouple/cylinder heater probes to get the temps to read right. All things being equal in the mean time, it isn't so hard to add 100 to what the EDM 700 is reporting. My guess is that the ship's single CHT is correct, and I will use it as primary until I get some concrete data to prove otherwise.

Thanks for the recommendations!

Cheers,
Rick

[kortopates]

I need to drop the cowling and have a look to see what's what, but in my research I believe I may have uncovered another issue with my particular installation that I need to consider - I have a Tannis engine heating system installed. I'll confirm it when I inspect the engine, but my understanding is that the Tannis system uses the CHT bayonet positions for the Tannis combination cylinder heating/CHT thermocouple installs. They are supposed to be K-types thermocouples, but my airplane is a 1989 and I haven't reviewed the log books to see when the Tannis system and the EDM 700 were installed and what exactly is in the thermocouple wells. I have a sinking feeling that I'm looking at about 1.2 AMUs in CHT thermocouple/cylinder heater probes to get the temps to read right. All things being equal in the mean time, it isn't so hard to add 100 to what the EDM 700 is reporting. My guess is that the ship's single CHT is correct, and I will use it as primary until I get some concrete data to prove otherwise.

Thanks for the recommendations!

Cheers,

Rick

Frankly, I would suspect the less reliable OEM gauge more than the JPI but you do need to keep looking deeper.

But good news in that if you are referring to the ancient legacy JPI, those old units supported both J & K thermocouple types to be compatible with EI probes.

But you are correct, you really want the Tannis heater dual heater/CHT probe in a grounded K thermocouple configuration- but they make them in all kinds - so no telling what you have till you uncowl it and take a look.

But if your old JPI 700 doesn't record data for you to download I would first remedy that with either a $1K upgrade to the color EDM 830 display or just go for the certified 900 or 930 ( or another of your choice) because with a valuable engine like yours, you really ought to have a modern engine analyzer to help you protect it.

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.

One more advantage of going to a certified monitor like the 900 or 930 is that's the only way you can pull the OEM CHT sensor out and have tannis heater element probes in all 6 cyl's - which is what I have. But you may not care or need them since you didn't install them.

 

 

Sent from my iPhone using Tapatalk

[ArtVandelay]

when we upgraded the 700 to an 830 ... we replaced that old spark plug ring probe/sensor with one that piggybacked off the factory probe. Problem solved.

Do you have a part number for this, I only fine the normal or spark plug ring probes?

[kortopates]

Do you have a part number for this, I only fine the normal or spark plug ring probes?

Look for JPI Adapter probe. I recall two different part no. for different thread/diameter. Once you figure out which one you need, order it through Spruce since they discount them pretty good.


Sent from my iPhone using Tapatalk