Menu Sign In Contact FAQ
Welcome to our forums

Induction Leak Test – Seeking advice

Hello folks, I am looking for the collective wisdom and experience of this forum. My IO-540 on a TB20, runs with GAMI injectors. I am observing the GAMI spread to oscillate between 0.0 gals/hr (unbelievable good) and 1.3 gals/hr (very bad). Furthermore, the leanest cylinder keeps changing as well. The leanest cylinder is mostly #3, or #1, but sometimes it is another cylinder.
Upon contacting GAMI, they suspected an induction leak. So I carried out two induction leak tests at two different altitudes (results shown in image below). The induction leak test is carried out at full throttle, and mixture full rich, recording the EGT temperatures, then throttling down to around 19 manifold pressure leaving mixture unchanged at full rich, recording the EGTs again, and looking for a cylinder indicating a much higher EGT relative to the other cylinders. The idea is that at full throttle any induction leak should have no impact, but when throttling down, any leak will cause a leaner mixture, due to the lower ambient pressure, and hence the EGT of the cylinder closest to the leak is higher, unless the leak so far upstream, that more than one cylinder would be affected.

The data I recorded for the two test runs are as shown in the graph. The light grey background color show the EGT temperature difference. The following stands out:

  • The test results of both tests are consistent.
  • When throttling down, the EGT of all cylinders rises, except for #6, which goes down (red circle).
  • The difference in EGT rise is highest for #1 (22 and 25 degrees), but still not much higher than the other cylinders.
  • When throttling down, there is no single cylinder that shows a dramatically higher EGT relative to the others, but I am not sure what exactly to expect.
  • I am not sure how to interpret the fall in EGT for cyl #5 (red circle)

How do I interpret the results? What would be the expected outcome with no induction leak? Would EGT rise, when throttling down, at full rich? Thank you!

An interesting question.
Personally, before i go flying around burning expensive fuel to try to find an induction leak, i would enroll the services of a good motor mechanic who should have a smoke tester. It’s a device that injects very fine smoke into the intake system and it will be seen escaping from the tiniest leak. I use one all the time with excellent results.

Last Edited by Stickandrudderman at 06 Oct 15:17
Forever learning

Send it to GAMI and have them tell you, no?

Mainz (EDFZ), Germany

I would tend to go with Stick’n’rudders advise, those systems are designed for what your trying to do. I don’t have a clue if a leakage would give such a different reading.


before i go flying around burning expensive gas to find

I actually not flying around burning expensive gas. This test just takes 3 minutes, and can be easily done on a routine flight.

enroll the services of a good motor mechanic

Not the best advice. You better collect as much info and data as possible, prior to handing things over to an A&P. Remember, once you enroll the services of a good mechanic the clock starts ticking, and you turn into an ATM machine. Besides, it is very difficult to judge whether a mechanic is good or not.

go with Stick’n’rudders advise

Not sure what you mean. What is a stick’n’rudders advice wrt engines?

Send it to GAMI and have them tell you, no?

Before doing this I like to collect as much info and data as possible. I.e. if I have an induction leak, it would not help sending an injector to GAMI.

Last Edited by Lucius at 06 Oct 17:03

Use of a smoke tester, these are designed to use for leaking find.


OK Lucius, it’s your aeroplane and your money, find an A&P with a smoke tester.

Forever learning

I have just heard from an engine specialist in the USA that there is a better test:

The data then requires interpretation and that bit I am not sure about but GAMI are apparently familiar with this.

Shoreham EGKA, United Kingdom

Send the data Gami is what I think he meant. Email.

EGPE, United Kingdom

Thank you Peter! This is helpful.

14 Posts
Sign in to add your message

Back to Top