How to improve the reliability of the 6 Cylinder Jab

[JabSP6]

John

 

Thanks for the photo. This looks very similar to my cowl. Do you have the dimensions for the upgraded Oil cooler? I would like to check that i have the latest one available. As for the deflectors i know they used to have metal ones that used to sit directly on the top of the cylinders which were recommended to be removed but i haven't tried the deflectors below the cylinders. This is something that Jabiru have not suggested but another fellow aviator has suggested it to improve the consistancy in the temps over the whole barrel. Will keep you informed of my progress. i will be trying these things one at a time while monitoring temps so i know if they work or not.

 

Wags

 

 

[Guest JRMobile]

Hi Wags

 

I made some enquiries re the MGL TC-3 egt/cht gauge and need some info on the length of the probe leads supplied. According to the supplier the leads are 1100mm in length, did you find that this was long enough to reach the back of the panel or did you have to extend the leads.

 

If you had to extend the leads, what type of cable did you use.

 

Cheers

 

John

 

 

[JabSP6]

John

 

What i did was go to a computer store and bought an extension cable of the right size plug. I think from memory it was a 23 pin plug. cost around $15. This saved having to join all the wires at the back of the gauge behind the dash. The pin plugged straight into the gauge and i cut the other end off the extension lead and ran it thru a grommet in the firewall. I then soldered all the wires from all 12 sensors to the computer cable out in the engine bay and had heat shrink around every wire. This is quite a lenghtly process to make sure all the positive and negative wires are hooked up correctly. I then just looped the excess sensor cable in the engine bay to keep things tidy. I was told that as long as you don't cut the sensor wires the sensors will still read acurately. You can extend the wires but you can't shorten them. To test this i then checked the reading against the original CHT sensor under NO6 plug and the reading was identical.

 

Wags

 

 

[Guest JRMobile]

That sounds like a very neat solution. thanks for sharing, as they say there is more than one way to skin a cat.

 

Cheers John

 

 

[JabSP6]

John

 

Come to think of it i must have used at least a 25 pin plug. There are 12 sensors with a negative and positive for each sensor and i'm sure ther was at least 1 spare pin.

 

Anyway the blokes from MGL will soon clarify this for you.

 

Hope this helps mate.

 

Wags

 

 

[DPW]

Unbelievable!!!!

 

 

[Powerin]

Reading the Arion Lightning newsletter today. There's details of a guy who's made mods to the Jab 3300 airbox (just a splitter vane I think) and seen marked improvement in EGT spreads. He has graphed the before and after temps.

 

Details and pics here: http://www.flylightning.net/images/pdf/Newsletter%203-12.pdf near the end. This may be old news to most of you.

 

 

[Vev]

Hi Powerin,

 

Whilst the Arion chap may have improved things, I would be quite concerned about the EGT temps over 1250 deg F ... I think this fellow will see a number of heat related problems very soon.

 

I think the easier improvement on the 6 cylinder EGT spreads is to simply remove the manifold divider and replace it with a 12mm round rod drilled thru the centre to bolt back into the manifold ... job done!

 

However, don't take my work for it, call Don at Jab and ask him.

 

Cheers

 

Vev

 

 

[jetjr]

Vev, can you elaborate this a bit

 

Im assuming the plenum has to come apart, and carb off?

 

Dou you have to disconnect all the intake tube too?

 

Its a fairly big job Id have thought?

 

 

[Vev]

Hi Jetjr,

 

Spot on ... the plenum has to come out, split plenum and replace the divider.

 

However, it's not that big of a job.

 

Cheers

 

Vev

 

 

[Jaba-who]

Splitting plenum

 

Does anyone have a diagram of the plenum chamber and the aforementioned divider? Or a pointer to a diagram if it exists in one of the jabiru manuals. I have downloaded all the manuals etc but can't find a diagram of the internal workings.(only a couple of very limited external views)

 

John

 

 

[Vev]

Jaba-who,

 

I think the best I can suggest is the parts manuals ... I think the 2200 parts book is possibly the best diagram, albeit not that clear.

 

Cheers

 

Vev

 

 

[Jaba-who]

Vev,

 

Thanks for that. I didn't think to look in the 2200 manual - assumed it would be "same" as the 3300 manual. But that image is OK for what I needed. I just wanted something to visualize what the divider was like.

 

Thanks.

 

John

 

 

[deadstick]

Hi WagsI made some enquiries re the MGL TC-3 egt/cht gauge and need some info on the length of the probe leads supplied. According to the supplier the leads are 1100mm in length, did you find that this was long enough to reach the back of the panel or did you have to extend the leads.

If you had to extend the leads, what type of cable did you use.

 

Cheers

 

John

Gday John, I finally got around to flight testing my unit today well things are very interesting. I used a 25pin extension cable for a computer to wire it in the engine bay and once heat shrunk pulled it into the cockpit and tied it up. Now down to temps I am seeing 70 degrees between hottest and coldest CHT hot- #2 Cold #5 and the egt from number 2 reached 726c momentarily, I have no choice but to richen this beat up it has the latest jets in the carb and no vacumn leak that I can find to explain #2 hotness. I cooled it down considerably by applying carb heat, I will fit a lamda guage soon to see what the A/F ratio is like but hell why such a massive split in egt and CHT? Manifold runner lengths unequal?

 

 

[JabSP6]

Gday John, I finally got around to flight testing my unit today well things are very interesting. I used a 25pin extension cable for a computer to wire it in the engine bay and once heat shrunk pulled it into the cockpit and tied it up. Now down to temps I am seeing 70 degrees between hottest and coldest CHT hot- #2 Cold #5 and the egt from number 2 reached 726c momentarily, I have no choice but to richen this beat up it has the latest jets in the carb and no vacumn leak that I can find to explain #2 hotness. I cooled it down considerably by applying carb heat, I will fit a lamda guage soon to see what the A/F ratio is like but hell why such a massive split in egt and CHT? Manifold runner lengths unequal?

Deadstick.

 

I was also amazed at the hugh difference in the CHT's and the EGT's once I fitted the MGL sensors. I have spent countless hours changing the path for the air to cool the heads to achieve consistant CHT temps. I too saw temps above 700C with the standard jetting of 185 needle jet and 255 main jet. To get my EGT's below 700C on cruise i fitted a 190 neddle jet. I also had to increase the main jet to a 260 so that a WOT the temps were lower than at cruise. This seems to be working well with only a minor increase in fuel usage. My EGT's now average around 680C at cruise and 650C at WOT. I believe this is why there are so many problems with overheating. People can only see the CHT of cyl 6 and no EGT's. In my situation cyl 6 was the coolest but I also had 60 - 70 C difference in CHT temps.

 

Wags

 

 

[deadstick]

Thanks Wags,

 

I suspected this to be the case and was not one bit surprised, I ordered the jets from Jab today so it should be sorted this week, I would hate to see what the old economy tuning gave!

 

DK

 

 

[Guest Andys@coffs]

Deadstick.I was also amazed at the hugh difference in the CHT's and the EGT's once I fitted the MGL sensors. I have spent countless hours changing the path for the air to cool the heads to achieve consistant CHT temps. I too saw temps above 700C with the standard jetting of 185 needle jet and 255 main jet. To get my EGT's below 700C on cruise i fitted a 190 neddle jet. I also had to increase the main jet to a 260 so that a WOT the temps were lower than at cruise. This seems to be working well with only a minor increase in fuel usage. My EGT's now average around 680C at cruise and 650C at WOT. I believe this is why there are so many problems with overheating. People can only see the CHT of cyl 6 and no EGT's. In my situation cyl 6 was the coolest but I also had 60 - 70 C difference in CHT temps.

Wags

Gents, When I worked on Aircraft in the RAAF any test equipment we used had to be calibrated and have a traceable calibration life such that when a measurement is taken iaw published instructions, you know with a degree of certainty that what the instrument is telling you is the truth.

 

How do you know that these significant temperature differentials are true differentials and not all or partially attributable to instrument error?

 

I can think of a few ways to check and see that the probes and individual channels arent contributing to the delta's you are seeing but they either involve significant work and repeated testing, or the use of another instrument/thermocouple pair that are calibrated, both of which are a pain....still if the added gear is to be of any use you must be able to answer if the delta exists because of different temperatures, or instrument/thermocouple paths...

 

I note that there are a number of IR reading hand held temp sensors (with a laser dot to identify the point of measure) but I havent looked to see what accuracy they are really capable of and what calibration accuracy exists for new items, nor the spread of the circular measuring diameter as a function of distance from surface... Let alone how you would use them with the engine running at the speeds we would need them to run at...

 

Lastly Im sure if you google "extending the leads of a thermocouple" there are a number of rule about what material you can use because I seem to recall that plain old copper wire introduces errors as it becomes yet another thermocouple junction in the circuit due to differing metal compositions. A 25pin D connector will create the same issue and is one of the reasons that the cheap and basic EGT/CHT systems that come from the factory have the thermocouple wiring going all the way to the instrument... Now I cant be bothered researching this with google so it mnay be technically true, but practically irrelevant, or as I seem to recall the actual voltages produced by a thermocouple are relatively small in which case any connector losses will introduce significant errors. My experience tells me that I would avoid extending the thermocouple lead if at all possible and I would wire directly back to the instrument. I would not cut the lead if too long but would try and store it in such a way as to minimise any inductive coupling, noting that the extensive outer shielding substantially helps with this, but at the same time try and avoid running in parallel next to the EHT leads if at all possible for example, or next to the alternator lead back to the voltage regulator as another example.

 

Andy

 

 

[fly_tornado]

the simple way would be too connect your EGTs to a multimeter and heat them up. you could try a gas torch.

 

 

[deadstick]

Points noted Andy, and yes you can sustain error through extending the T-couple lead, this particular instrument states that it has an advance algorithm to combat this issue and I would never change the length of the lead (constant) and I have looped it under the dash. The extension required is in mine 20 CM and given that it is high quality wire I dont see too much of a potential loss. As The constant (T-couple) and extension are the same length for all any error would be the same across the board, Yeah? I check calibrated the CHT T-couples before fitment and they were right on, in fact the ambient temp this morning was 25c and all chanells displayed this. Remembering that Jabiru use a fairly rudimentary CHT T-couple and no it does not wire all the way to the guage its 25cm long and extends with ordinary cable. I have used my Laser temp gun to confirm what I am seeing on the ground. Heres what MGL have to say:-

 

 

 

Thermocouple leads as used with the EGT and CHT probes can be extended either with ordinary copper cable or with

 

 

 

 

 



special K-Type extension cable. The choice of either depends on your desired accuracy. If it is possible in your installation

 

 

 

 

 



to ensure that both ends of a copper extension cable will be at the same temperature (or very close), then it is quite

 

 

 

 

 



possible to use the copper cable. In most open-air installations this will be the case. Should this not be possible or you

 

 

 

 

 



require best possible accuracy at all times, you can obtain a special K-type extension cable. This cable is made from the

 

 

 

 

 



same metals as your probes cable and uses ordinary plastic sleeving as insulation. In either case, ensure that the cable is

 

 

 

 

 



not routed close to sources of electromagnetic interference of any kind. The voltages present in this cable are very small

 

 

 

 

 



and are subject to changes applied by external fields. This can lead to false temperature indications. You can check your

 

 

 

 

 



installation by using a hand-held transmitter, such as an air band radio. If you transmit a signal, no change in temperature

 

 

 

 

 

reading should occur.

 

 

 

I think the most important thing is to make sure the instrument is earthed the same as the engine. And further more the engine has displayed signs that what the guage is saying is right with #2 running hot.

 

 

 

[JabSP6]

How do you know that these significant temperature differentials are true differentials and not all or partially attributable to instrument error?Andy

Andy

 

The first thing i did when I discovered the differences in CHT's was to swap the position of the sensors to make sure the Temp was in fact correct for that head and not follow the sensor indicating a sensor fault. All was good to my surprise. No matter which position I fitted the sensors the CHT's always remaind the same for the head. As all i'm trying to do is reduce the temp as much as possible and even them all out i'm not that worried if all sensors are not exactly calibrated perfectly. One thing is for sure. I have managed to reduce the head temps on my 3300 from averaging 160 - 170c at WOT and 140 - 150c at cruise to a much more pleasing 115c at WOT and 105c at cruise (average). If these actual figures are 5c hotter or lower doesn't really matter to me. The main thing is i have reduced and been able to even out the temps by fitting sensors to each cylinder and modifying the air flow to each head, and hopefully do what this thread was all about. "Help improve the reliability of the 6 cylinder Jab".

 

Wags

 

 

[Guest Andys@coffs]

Wags

 

Accept all that, my comments were just to make people consider that there may be multiple aspects to the problem. Murphys law says if you dont consider it, it will end up being significant.

 

If you have head temps of 105 at cruise, what oil temps do you get? Is it high enough to boil off moisture? (Ironic that I ask that Q about a jab of all things!!) It wasnt that long ago that I fitted the largest of the oil radiators to my J230 which made a difference in hoter weather, where as before I might not have been able to climb above 2000ft on a stinker of a day (back when I was in SA) I now can climb up to cooler temps.

 

Andy

 

 

[facthunter]

wags,transposing the sensors covers all that. (as you have done) The temps you quote are surely cold enough, In fact I would say you are colder than necessary, but you won't do much damage being a bit cool. (In the short term).

 

I am sure jabiru have been working on getting fuel mixture balance with the plenum design for years. It's a nearly impossible task. When I was racing ( mainly building engines) I would never use a siamesed port engine. (Inlet) and preferably not in the exhaust either. Most were one carb to one cylinder. This makes the whole thing possible and is actually not very difficult to tune. This cannot be done on the engine we are talking about here. I think that what you are taking on is beyond the resources of most individuals., and even most companies. Your EGT measurements will give you a good indication of mixtures but how to correct a detected fault I wouldn't guess . You have pulsing, pressure waves, bends, lengths all influencing the outcome. Nev

 

 

[Guest Andys@coffs]

.....in fact the ambient temp this morning was 25c and all chanells displayed this. Remembering that Jabiru use a fairly rudimentary CHT T-couple and no it does not wire all the way to the guage its 25cm long and extends with ordinary cable. I have used my Laser temp gun to confirm what I am seeing on the ground. .......

So I can see the relevance of the ambient temp to the Cyl Head sensors But I would have thought that the EGT sensors operated in a temp range that had 25deg C well outside their usefull measuring range.

 

In any event you clearly were aware of all the points I raised. Just out of interest the laser pointed IR sensing instrument, what brand is it and what is its stated accuracy? (If you have the details at hand) I quite like the idea of those but wonder how you could use one when engine operating as intended given the fibreglass shrouding around the heads for CHT and as to EGT, not sure how you measure that one given the thermocouple extends to the middle(ish) of the gas flow where the IR sensor I pressume can only measure the pipe surface. I pressume (Without any knowledge on which to base the presumption) that the surface temp and the centre of the pipe gas temp are very different...

 

Andy

 

P.S in my J230 the basic 2 channel CHT/EGT I'm pretty sure (well as sure as I can be when its about 70k's away in the hagar waiting for the flood to come...) has thermocouple cable all the way to the sensor.....that said, I can well imagine that what ever is present from shipment to shipment will depend on which ever Chinese knock off was sourced from one shipment to the next.. To me that system as it is, is merely there to fill a whole in the panel that would otherwise be a talking point.....

 

 

[deadstick]

So I can see the relevance of the ambient temp to the Cyl Head sensors But I would have thought that the EGT sensors operated in a temp range that had 25deg C well outside their usefull measuring range.I have used K type thermocouples and As you would have learnt at RAAF STT, quote- K type thermocouple is the most popular and uses nickel-chromium and nickel-aluminium alloys to generate voltage.Standard tables show the voltage produced by thermocouples at any given temperature, so the K type thermocouple at 300°C will produce 12.2mV end quote....

 

Makes little difference if its for CHT or EGT, voltage is known for heat range the only real variable is the extension and the Junction and you must be sure the instrument is at the same potential as the lead.

 

In any event you clearly were aware of all the points I raised. Just out of interest the laser pointed IR sensing instrument, what brand is it and what is its stated accuracy? (If you have the details at hand)

 

I don't have the details at hand but have it sitting here it is a kingcrome unit and distance held when measured was around 20-30 cm from the target.

 

I quite like the idea of those but wonder how you could use one when engine operating as intended given the fibreglass shrouding around the heads for CHT and as to EGT, not sure how you measure that one given the thermocouple extends to the middle(ish) of the gas flow where the IR sensor I pressume can only measure the pipe surface.

 

Ahh young padawaan 'answer seek you must!', for EGT loosen the clamp and allow some of the probe to be exposed and as for CHT take the cowls off, LOL. Both were measured during ground run and were accurate, would still be happy if I had a 5% error.

 

I pressume (Without any knowledge on which to base the presumption) that the surface temp and the centre of the pipe gas temp are very different...

 

Andy

 

P.S in my J230 the basic 2 channel CHT/EGT I'm pretty sure (well as sure as I can be when its about 70k's away in the hagar waiting for the flood to come...) has thermocouple cable all the way to the sensor.....that said, I can well imagine that what ever is present from shipment to shipment will depend on which ever Chinese knock off was sourced from one shipment to the next.. To me that system as it is, is merely there to fill a whole in the panel that would otherwise be a talking point.....

Yours has a 2 channel unit, my 230 came with 1 channel (ripped):thumb_down:

 

Next I will be fitting a lamda guage.

 

 

[fly_tornado]

what about a knock sensor?