GAMI spread

[FloridaMan]

I have started messing around with my new EDM-900. Even though my plane is equipped with GAMIjectors, I'm thinking that they may not be balanced correctly. When going through the "ROP" procedure in the manual in cruise at < 75% power, I find the "leanest" cylinder and then enrichen to say, 50 degrees ROP. I notice that when the display is cycling through the readings that it shows cylinders 2 and 4 with a difference of 80-100 degrees. I was under the impression that GAMIs were supposed to be closer to that. I also have tried the LOP procedure, but it seems like the engine likes to stumble after finding the richest cylinder. 

[Vance Harral]

GAMI spread has nothing to do with different EGT numbers between cylinders.  It has to do with the fuel flow at which each cylinder reaches its individual peak EGT.  In a well balanced engine with GAMIjectors, all cylinders should reach their peak EGTs within a few tenths of a gallon-per-hour as you lean the mixture.  The value of peak EGT in any one cylinder may differ by 100 degrees from peak EGT in some other cylinder, but that is "uninteresting".

[Houman]

you can actually call GAMI up and the guys there are pretty good at going trough their stuff and how to calculate your gami spread.

I did mine and found out that my GAMI spread was too low to require any new work to be done on it..so that was a good thing.

[carusoam]

Antares,

Are you able to download the data from your JPI?

Going through the ROP and LOP processes the first time using the JPI I surprised myself at how complex and challenging it is to get it right...

I found things like having it set up ROP while running LOP and vice versa... (Setting the mode properly)

The 900 must be better than the 700 for this, but they probably have the same confusing challenge.

At least downloading the data, you can review the steps you took and the results you got to see if it makes sense. Having FF numbers associated with the EGTs is real engineering data excitement!

Trying to do this while flying will take longer...

Is this correct...?

LOP uses last to peak

ROP uses first to peak

Best regards,

-a-

[Shadrach]

Disregard raw numbers and look at spread between peaks.  A hypothetical example:

 

Cruising at t 7500

 

Cyl 1 - Peaks at 1460df and 10.2 GPH

Cyl 2   Peaks at 1535df and 10.3 GPH

Cyl 3   Peaks at 1575df and 10.1 GPH

Cyl 4   Peaks at 1498df and 10.2 GPH

 

The Gami spread here is .2 GPH which is pretty darn good.

The EGT spread is 115df which is...no reason to give a $hit!

 

IMHO EGT raw numbers are really only important under one scenario - Mag/plug failure.  North of 1600 in cruise on a NA engine usually means only one plug is firing.

[Awful_Charlie]

EGT is *very* dependent on how far the probe is from the valve, so unless the probes are exactly the same distance, with the tip in the hot part of the flow (more difficult) then even in the case that the cylinders are exactly the same, then you will get a difference in the recorded/displayed values.  The actual number of the EGT is useless, is is the difference from a known number that is useful

[Bob_Belville]

I talked to the GAMI folks at SunNFun a couple of year ago. They were very helpful and suggested that our IO360 usually does not need GAMI injectors. My JPI 930 confirms that assessment, My spread is usually 0.2 gph or less. I download and check data with EZTrends2 after every cross country flight. (I also load to Savvy's site, that software has some nice features but the LF function of EZT is slick.)

[Shadrach]

I agree Bob, almost every NA IO360 Ive known has excelent distribution.  A lot of it is due to the intake design (evenflow to each cylinder). Some of the newer engines are not so great.  I wonder if there are casting imperfections in the airbox on some of the newer engines that are disrupting flow.

 

My A1A has stock injectors and will easily run 70 LOP at 2500rpm. 

[Bob_Belville]

I agree Bob, almost every NA IO360 Ive known has excelent distribution.  A lot of it is due to the intake design (evenflow to each cylinder). Some of the newer engines are not so great.  I wonder if there are casting imperfections in the airbox on some of the newer engines that are disrupting flow.

 

My A1A has stock injectors and will easily run 70 LOP at 2500rpm. 

I might be able to lean that far but power and speed drops off pretty steeply that far lean of peak. I think Jetdriven has quantified and optimized the best point on the curve for economy.

[Shadrach]

I might be able to lean that far but power and speed drops off pretty steeply that far lean of peak. I think Jetdriven has quantified and optimized the best point on the curve for economy.

 

 

I'm not in any way suggesting that 70 LOP is optimal. I was using it to illustrate that I have a well conforming engine. I find that for my IO360 the most optimal combination of cylinder temp, speed  and fuel burn occur between peak and 40 LOP depending on DA.  

[Txbyker]

A quick little trick I have used is to film the leaning with your smartphone then when at a computer watch the leaning frame by frame.  Very easy to determine first and last to peak and thus the spread.

 

Russ

[jetdriven]

A quick little trick I have used is to film the leaning with your smartphone then when at a computer watch the leaning frame by frame.  Very easy to determine first and last to peak and thus the spread.

 

Russ

That's a great idea.

[FloridaMan]

I'm not sure if I did things right, but I went up to 7500MSL with the ram air open (makes a difference with the lopresti cowling - ~2 inches). I selected "Lean Find" in the EDM-900 and set the mode to LOP so it would show me "Leanest" and then "Richest" cylinders. At 7500+ram air, it started flashing "Leanest" at 10.1 GPH and "Richest" at 9.1GPH and the engine started running rough. 

 

At 6500MSL with the ram air closed, it showed "LEANEST" at around 9GPH and I leaned it to < 8GPH, the engine started running rough and the EDM-900 never got to "RICHEST." 

[Bob_Belville]

at 1.0 gpm gami spread I would say you'll need to work on balancing the injectors closer in order to run efficiently. From what we've read here often IO-360 should be more like 0.2 gph spread.

 

There could be several causes and solutions.   

[Shadrach]

I am betting that you are leaning too fast.  Many folks do this when the do their first lean test because they fear damaging their engine. At 7500ft in FL, its highly unlikely that  you could damage your engine no matter what combination of throttle and mixture you set.   

 

Now that you know (hopefully) which is leanest and richest. The next time you go up, take your time finding the peak for both of those cylinders; this means waiting a 4 or 5 seconds after you think you're at peak and then fine tuning to find absolute peak EGT. It may take 20 to 30 secs for each Cylinder for you to find true peak. Note the fuel flow at each cylinder's peak; that is your GAMI spread.  My A1A's leanest cylinder is #3 (used to set ROP) and the richest cylinder is #2 (used for LOP).  I have not yet installed FF so I don't know my spread, I do know that I can smoothly run way further LOP than would be useful and that's good enough for me.  

[Bob_Belville]

I note from your gallery that you have a mixture control with verier, if you are following the JPI LF you are gradually, steadily, backing out the mixture. My guess is your technique is not the problem but if Ross's shoe fits...

[cujet]

Good info above. 

 

lean slowly, follow the GAMI procedure, look at the downloads. 

 

My IO-360 with Gami Injectors has roughly a 0.3 GPH spread between the first and last cylinder to peak. I run 25 degrees timing and lean stumble comes well past 100 degrees LOP. I'm sure I could "tweak" it so they all peak at very nearly the same time. But, it runs smooth and strong, so no real need. 

 

Also note that 20 degrees timing may result in rougher, or at least less pleasant, LOP operations. As lean mixtures burn more slowly. The 38 degree timing advance of electronic ign works wonders during LOP operations at high altitude or low manifold pressures (they are one and the same) 

[Shadrach]

You're right on! 20BTDC does not make for great ops LOP. Combustion event still taking place on the exhaust stroke. Increased EGTs decrease in performance. Almost like running on one Mag at 25BTDC.   I think the optional 20 degree timing set up is a bad deal.  I think it will help reduce temps in a bird that has poor baffling or some other cooling deficiency.  There are loads of IO 360s operating well within parameters at 25BTDC in a lot of different applications. what is different about the engines that run hot and those that don't?  I have to believe it's the cowling, baffle material or installation. I would detail my the flow through the engine compartment long before I'd retard the timing.