Cat D5G service manual and help

[Snapper]

Dozer starts and works great, intermittently the parking brake light comes on and one track stops. Yes we have tested pressures and get 350 psi for parking brake. We have checked codes and have several intermittent voltage codes for parking brake and frequency of motors. We have checked the harness, alternator, batteries ect. I'm asking to see if anyone has a manual for this machine. otherwise if anyone has input on things to try. One thing i have found is on the parking brake switch, Depending on orientation of the switch 12v comes in on the back leg or the front, testing continuity of the switch it shows each leg should correspond to the center. test the center with power probe it trips the breaker when 12v is applied. using base knowledge without proper schematics I'm looking to replace the ecu. I just need to be sure prior to doing so.

 

[Nige]

Based on your reported symptoms of intermittent Code the first thing I suggest is to carefully check the condtion and tension of the fan belt & pulleys. This model is notorious for under-voltage issues if either the belt and/or pulleys is in poor condition or not tensioned correctly, or both.

One thing i have found is on the parking brake switch, Depending on orientation of the switch 12v comes in on the back leg or the front, testing continuity of the switch it shows each leg should correspond to the center. test the center with power probe it trips the breaker when 12v is applied. using base knowledge without proper schematics I'm looking to replace the ecu. I just need to be sure prior to doing so.

If you would like an electrical schematic then please post a machine Serial Number.
Same comment applies in order to identify the appropriate Service Manual for the machine.

 

[Snapper]

Based on your reported symptoms of intermittent Code the first thing I suggest is to carefully check the condtion and tension of the fan belt & pulleys. This model is notorious for under-voltage issues if either the belt and/or pulleys is in poor condition or not tensioned correctly, or both.


If you would like an electrical schematic then please post a machine Serial Number.
Same comment applies in order to identify the appropriate Service Manual for the machine.

I have a photo of machine serial and model in original post, I will also include it here. model D5G SN CAT00D5GPFDW00885.

We have checked alt and belt.

 

[Nige]

We have checked alt and belt.

Try putting a bit more tension on the belt. You have nothing to lose at this point. Correctly tensioned it should be almost bow-string tight.

Sorry I'm so used to people not posting Serial Numbers that the line "post a Serial Number" comes out automatically.

Have you tried erasing all the Codes to see which ones come back.? Note: You cannot erase anything that is Active.

Check your message inbox for an electrical schematic.

A full Service Manual for your Serial Number is Caterpillar Publication reference RENR5510

 

[Snapper]

Try putting a bit more tension on the belt. You have nothing to lose at this point. Correctly tensioned it should be almost bow-string tight.

Have you tried erasing all the Codes to see which ones come back.? Note: You cannot erase anything that is Active.

Check your message inbox for an electrical schematic.

yes we have erased and produced new codes and a couple same old ones, nothing is active, only history codes after running for an hour.

 

[Nige]

Can you supply a list of the history Codes please.

Are you using a scanner/ET to communicate with the machine electronics or are you pulling the Codes by disconnecting the Harness Code plug.?

 

[Snapper]

Can you supply a list of the history Codes please.

Are you using a scanner/ET to communicate with the machine electronics or are you pulling the Codes by disconnecting the Harness Code plug.?

using a generic scanner. codes are as follows:
70-2
190-8
153-8
154-8
444-5
133-5
1706-5
1704-5

 

[Nige]

using base knowledge without proper schematics I'm looking to replace the ecu. I just need to be sure prior to doing so.

Personally speaking I think it is unlikely that the ECM has failed although it is a possibility.

Based on your list I would still put money on the alternator output not being sufficiently stable over time.

only history codes after running for an hour.

If your scanner can satisfactorily erase all the historical Codes then I would suggest that you do that and instead of operating the machine for an hour do it for just 5 minutes then see what has come back.

If you want to be sure that the Codes are definitely erased you can use a jumper wire across the Harness Code connector which is #2 on the illustration below. Simply unplug the two halves of the connector, turn the key ON, and use a jumper between Pins 2 & 3 of the machine harness connector to erase each history code one by one. The jumper will need to be removed and reinserted after each Code is cleared. At the end of the process the Blink code on the indicator #4 should be 222 if all the history codes were erased satisfactorily.

 

[Snapper]

If your scanner can satisfactorily erase all the historical Codes then I would suggest that you do that and instead of operating the machine for an hour do it for just 5 minutes then see what has come back.

If checking after 5 mins there may be one of those codes and there may not. Very random.
The largest of the concerns is that when pushing a large load one or both tracks may stop and parking light is intermittently coming on. We did notice when this occurs, parking brake pressure is showing 350 and randomly drops to 240. if we idle down pressure goes up, idle up pressure goes down. shut system down toggle the brake switch and start back up and its normal until it gets under a heavy load.

 

[Nige]

We did notice when this occurs, parking brake pressure is showing 350 and randomly drops to 240. if we idle down pressure goes up, idle up pressure goes down. shut system down toggle the brake switch and start back up and its normal until it gets under a heavy load.

It might be interesting to monitor the output voltage from the alternator while all this is going on.

The Codes on your list that most closely seem to be related to the symptoms are: -

1. 133-5 which is the Transmission Pilot Pressure Solenoid Valve. The -5 indicates that the current in the circuit is below normal or the circuit is open. I would suggest that in addition to the parking brake pressure you that monitor both front and rear pump charge pressure on Pressure Taps #3 & #4 while operating the machine.

2. 70-2 which is the parking brake Rocker Switch. The -2 indicates that the data coming from the switch is intermittent, invalid, or incorrect.

 

[Nige]

using base knowledge without proper schematics I'm looking to replace the ecu. I just need to be sure prior to doing so.

Just came to me - one Code that is not even in the historical list is any variation of 0041-x which indicates a failure in the 8v sensor supply from the ECM. That could possibly be an indication of an ECM problem but, as it is not in the list, it isn't.

 

[Snapper]

2. 70-2 which is the parking brake Rocker Switch. The -2 indicates that the data coming from the switch is intermittent, invalid, or incorrect.

With that being said measuring voltage coming into the switch it’s 12v coming out of the switch it’s grounded going back to the ecu.

 

[Nige]

Can you indicate on which one of the wires in the illustration below you were measuring 12v.?



Parking Brake Switch logic.

The parking brake switch has a normally open contact (#3) and a normally closed contact (#1).
The contacts provide switch to ground inputs to the ECM in order to indicate the position of the switch.
When the ECM determines that the parking brake switch has been engaged and the machine is in motion it will fully upstroke the motors and will fully destroke the pumps, then apply the parking brake.
The following table provides the switch logic for the parking brake switch.

 

[Snapper]

Can you indicate on which one of the wires in the illustration below you were measuring 12v.?

View attachment 351664

Parking Brake Switch logic.

The parking brake switch has a normally open contact (#3) and a normally closed contact (#1).
The contacts provide switch to ground inputs to the ECM in order to indicate the position of the switch.
When the ECM determines that the parking brake switch has been engaged and the machine is in motion it will fully upstroke the motors and will fully destroke the pumps, then apply the parking brake.
The following table provides the switch logic for the parking brake switch.
View attachment 351665

1 and 3 have 12v key on and 2 has ground.

 

[Nige]

1 and 3 have 12v key on and 2 has ground.

That is strange. I would have expected a pull-up voltage maybe somewhere around 5v. The fact that there are no Active Codes while performing this test is also indicative of the circuit performing as it should do.

All that Pins 1 & 3 are supposed to do is provide a path to ground (actually back to the ECM) via wire J764. If you look at the ECM connector J1 Pin 12 on the schematic it is indicated as "Track Speed Sensor RTN" but in actualy fact it is a ground/return connection internal to the ECM for a bunch of switches/sensors. Chase all the branches out on the schematic.

Basically the ECM determines the position of the parking brake switch dependent on whether it detects ground on either Pin 1 (wire A314 Orange) or Pin 3 (wire A315 White) of the parking brake switch.



Have you taken a good look at the two 70-pin connectors on the ECM for corrosion or green crusties.?

 

[Snapper]

Have you taken a good look at the two 70-pin connectors on the ECM for corrosion or green crusties.?

 

[Snapper]

Have you taken a good look at the two 70-pin connectors on the ECM for corrosion or green crusties.?

Sorry, I thought I had already replied to this. but yes connectors have been check and they are in good condition. no corrosion or any type of obstruction.

 

[Nige]

Sorry, I thought I had already replied to this. but yes connectors have been check and they are in good condition. no corrosion or any type of obstruction.

In that case I would suggest to connect a multimeter to the park brake solenoid and pressure gauges on the parking brake test point plus Tests points 3 & 4 (front & rear pump).
Run the machine like that and see what happens when it acts up.

 

[Simon C]

Personally speaking I think it is unlikely that the ECM has failed although it is a possibility.

Based on your list I would still put money on the alternator output not being sufficiently stable over time.


If your scanner can satisfactorily erase all the historical Codes then I would suggest that you do that and instead of operating the machine for an hour do it for just 5 minutes then see what has come back.

If you want to be sure that the Codes are definitely erased you can use a jumper wire across the Harness Code connector which is #2 on the illustration below. Simply unplug the two halves of the connector, turn the key ON, and use a jumper between Pins 2 & 3 of the machine harness connector to erase each history code one by one. The jumper will need to be removed and reinserted after each Code is cleared. At the end of the process the Blink code on the indicator #4 should be 222 if all the history codes were erased satisfactorily.


View attachment 351624

Have tested two different machines lately that had too much Alternator AC ripple voltage and it was causing wonky things. On one the computer was reporting voltage problem.
Very fast test to do to rule out defective diodes causing voltage and amp problems.
Multimeter on AC Volts not DC. Red test lead on output post of alternator with engine revved up and lights and fan on to create load, while putting Black test lead on Negative post of Battery. A reading over 500MV is indicative of faulty diodes in alternator on a 12 volt system. I don't have the schematic but if it is 24 volt then the number would be like 1.0 volt AC.
Worth testing and may help in diagnostics before buying Alternator.
Simon C

 

[Nige]

It certainly would be cheaper than investing in a replacement ECM which is what the OP was thinking.