New member

Re: combustion sensors/burn times

Los **** indican que el valor leido es demasiado grande, o fuera de rango, lo que normalmente ocurre en sensores de temperatura.

En un sensor de combustion, solo debe tomarse en cuenta cuando el equipo esta en funcionamiento con carga.

Cuando indica que existe un problema intermitente, es posible que este ocurriendo los siguientes problemas:

1. Sensor defectuoso, debe medir menos de 1 ohm

2. Entre cables y conectores debe medir menos de 1 ohm

Finalmente al instalar el sensor hay que ser cuidadoso, porque puede no encajar correctamente en el orificio del conector, de tal manera que internamente no se ve que esta mal conectado pero la vibracion hace que el contacto sea erratico.


Lo que yo hago normalmente es, desconectar el cable medir.

Sacar el conector y verificar que el extensor calse apretado con la punta del extensor para cuando exista vibracion no sea afectado.




The **** indicate that the value read is too large, or out of range, which normally occurs in temperature sensors.

In a combustion sensor, it should only be taken into account when the equipment is operating under load.

When it indicates that there is an intermittent problem, it is possible that the following problems are occurring:

1. Defective sensor, must measure less than 1 ohm

2. Between cables and connectors must measure less than 1 ohm

Finally when installing the sensor you have to be careful, because it may not fit correctly in the hole of the connector, so that internally it is not seen that it is badly connected but the vibration makes the contact erratic.


What I usually do is disconnect the cable measure.

Remove the connector and verify that the extender is tight with the tip of the extender for vibration is not affected.




Re: combustion sensors/burn times

It has been awhile and we still have not solved this issue. we had a training class for tuning the 3600 and I was not invited to the party. hate to say it but it has left a little bit of a salty taste in my mouth. today, I am investigating more engines that are having the same issue and no apparent reason why. I have not forgotten about this thread and if I happen to stumble upon the fix, I will be sure to respond here in hopes of helping someone else.


If any of you happen to have different ideas, I am all ears!!! I can completely understand how difficult it can be to diagnose something without laying eyes and hands on it personally. I appreciate all the help so far.


Re: combustion sensors/burn times

After a bunch of communicating through email, we have still not determined our issue yet. Al has poiinted out that we have not been doing the entire tuning process and it MIGHT be contributing. That is going to be our next step is to follow the tuning process in the CAT manual sentence by sentence and see if that helps us.


The data logs are showing almost everything happening at the exact same time ( engine drop, throttle actuate, combustion data break in connection on cyl 1) and then the wastegate reacting after the event.

New member

Re: combustion sensors/burn times

So what was the reason? We have same problems! 


Re: combustion sensors/burn times

sent you an email with the attached datalog file.


Re: combustion sensors/burn times

Thank you for taking this time to help us out!


I will get that today. This first datalog was more of an experiment and I was trying to think of what else I could log to find the trigger. The log I ran was only 5 min and it caputured at least 4 stumbles.

Trusted Contributor
Trusted Contributor

Re: combustion sensors/burn times

Thanks for the new information. However, we need to do this a bit differently so that I can really evaluate what the engine is telling us.  We also need a few other items of information in the data log.


If you will, set up a data log of these items:

1. Desired engine speed

2. Engine speed

3. Desired Air Pressure

4. Actual Air Pressure

5. Air to Fuel Differential Pressure

6.Desired Intake Manifold Air Pressure

7. Engine Load Factor

8.Fuel Temperature

9.Hydrax Oil Pressure Switch Status

10.Per Cent Fuel Position

11.Wastegate Position Command

12 Wastegate Position

13.Choke Position Command

14.Actual Engine Timing

15..Gas fuel Correction Factor

16.Fuel Valve Position

17.Desired Air to Fuel Ratio

18.Actual Air to Fuel Ratio

19.Choke Position Command

20.Actual Ignition Timing

These items will give a more complete picture of what the ECM is doing and how the engine is reacting.

I don't know how long you have to record a data log to get an engine hiccup but if you can catch the engine in another upset with this new data log parameter group at 240 samples a minute, we might be able to zero in on the cause of the upset.

Send the data log file to me at and I will take a look at it on ET so that I can manipulate the data for several comparisons.

After I have had an opportunity to review the data log, I will respond back to you.  I really don't have enough data yet to have an idea of what is going on.  Thanks.

Best Regards,

Al Hunt


Re: combustion sensors/burn times

[ Edited ]

So, I was hoping to continue this conversation but maybe twist it a little bit. I was able to do a little datalogging on this unit that seems to have a stumble. This is not the only unit that is doing this. Out of 25 G3606's there are probabaly 10 or more with a similar issue but none quite as much as this unit.


A fellow coworker has also been messing with the same unit under the guidance of a local CAT dealer. we still have yet to figure out the issue. This particular unit will be running right along at 1,000 RPM and at the snap of the fingers, it will drop to 985 RPM. Then, sometimes just as quick but often a slower climb, it will regain the lost rpm and even go over up to 1,010 or so. All combustion sensors have been replaced and a couple suspect data cables. Also, the Fuel/Air module was recalibrated and the throttle and wastegate acutuators were replaced along with several hydrax fluid changes. The poor running still exists.


The datalog that I recorded was of desired engine speed, actual engine speed, all 6 unfiltered combustion times, fuel throttle percentage postion, and wastegate percentage position. When looking though the log, I have noticed that the only thing consistant is the RPM drop sharply at the same exact time the throttle increases sharply and at least one cyl burn time goes blank. Most of the drops were showing cyl 1 combustion data disappearing from the datalog then reappearing a fraction of a sec later. I have a screen shot of one of the drops in engine RPM that I will hopefully get attached to this.


Is there something in this system that could possibly cause a run time distrubance and make the engine stumble? It is almost like the loss of the compustion signal is directly linked but I don't know how. Maybe whatever is causing the loss of the combustion signal is also responisble for the engine stumble.


the engine serial number is 4ZS01788


Thanks for any ideas on this issue.


ac4497 datalog.png


Re: combustion sensors/burn times

Thank you sir, that is greatly appreciated. We have ICSM if I remember correctly but will verify to be sure.
Trusted Contributor
Trusted Contributor

Re: combustion sensors/burn times

The ***** indicate that the ECM is not receiving a signal from the combustion sensor from the cylinder that shows the stars.  This can be caused by several things. 

One of the most common reasons for this event is moisture affecting the combustion feedback cable.  Either moisture has seeped into the coaxial cable assemby or there might be moisture at the connection on either end.  This might be supported by your comment that more cylinders show this when the engine has been started after not running for a period of time.


I don't know where you are located but a high humid climate can increase this opportunity for occurrence.


Or it could be that the connections are just loose on either end.  It could be that the cables are losing their integrity and getting close to needing replacement. 


It could be the ISM, if that is what is on your G3616.  Your serial number would indicate which component, the ICSM or the ISM, is on your engine.  The early software on the ISM also caused some irregular **** events.  Cat started using the ISM with G3616, s/n BLB00691.  The later Flash File to fix this issue in the ISM is 451-5846.


It could also be that the software in the remote advisor needs to be updated.  This has caused **** to appear when they should not.  You might need in the advisor Flash File 468-9789 to fix this issue.  You can use ET to see what FF is in the advisor.


So you have several things to look at.


Best Regards,

Al Hunt