Hi people, I am having an issue with a 320D2 (Serial Nº PFE10094). The machine was working 3 weeks ago with no issue and yesterday it didn´t start. Not even crank. The display turns on with the key, but there is no info (like fuel, temperatures, etc). We tried disconnecting the ECU (part Nº 386-3444) and it is the same. There are no codes also in the dash, like there is no comunication with the ECU. In one corner of the ECU, there is a red led blinking when the ignition is on. We tried to start the engine bridging (idk if it is the corret word) the starter motor and the engine cranks well and even start for a few seconds. All fuses are ok, and we tried switching places of the relays and didn´t change anything. We measure with multimeter the red wires in the connector of that ECU and there are 24v and Ground. I test some sensors for 5v but none of them have steady 5v, some have some come-and-go 3.5, 0.8 and so. I have a 320DL (Serial Nº A8F00985), I am not sure if the ECUs are the same and I can test switching them. I don´t want to burn another ECU if they are not compatible. If anyone have any information it will be very appreciate.
Has any welding been done on the machine recently.? Check your message inbox. I will send you an electrical schematic so that you know what you are testing. The 5v sensor signal wires will generally not show 5v, especially if they are the PWM type. In that case you will require a multimeter capable of measuring duty cycle to check sensor output. The Machine ECM for machine Serial Number A8F00985 is not the same Part Number as the machine with the fault. I would not recommend switching the ECMs even for a limited test. It may be necessary to connect Cat ET to the machine in order to make a full diagnosis.
Hello Nige, I just came from the machine. No, there has been no welding in that machine. A few weeks ago, there was a thunderstorm near, maybe I am very unlucky. I checked the machine with the electric schematic you send me, there is 24v in + pins both of Machine and Engine ECMs. I also check and there is 24v in the +Key Pin of both ecus. I connected the ET but I can only see the monitor system. I didn´t check sensor outputs, but I am starting to be more sure that maybe it is a ECU problem. Is there a way to test if the ECUs are working?
Not really. If ET will not communicate with a particular ECM there is a possibility that re-flashing the software will bring it back to life, but it’s not guaranteed. That process will require a dealer version of ET.
Hello people. I am with (almost) the same issue as before. I dismantle all the cabin to see how where the cables and some white plastic connectors where really bad. I bypassed the connectors and now the machine is starting well, but still have no info in the monitor. Also, the manual accelerator of the cabin is not working. The hydraulics of the machine works, but as there is no accelerator, when you try to move anything the engine stops. I was reading some older posts similar to this and maybe I am ponting to the cabin main fuse or main relay. It is weird becasue the monitor is turned on, but no info is displayed. Both the CAN and RX cables have continuity from the monitor to the ECM and have 60ohms.
ET will show you which ECM's are communicating. Determine if it's communicating first and we'll go from there.
Just came back from machine. Both in the monitor and with the ET there is no communication with ECMs. There is continuity in both of the can lines, and the resistance is 60ohms. The voltage of the CAN (J1939) is 1.6 and 4.5 with key on. On the machine CAN (EDCM), there is 4.5 volts on both canh and canl with the key on. I don´t know if that´s ok. Also, I was measuring the resistance between can pins and ground. I have around 150/300 ohms in both the can lines. From what I understand, I maybe have a shortcut to ground somewhere. I dismantle all the plastics of the cabin and take all the cables apart, all seem ok. The machine starts well, but there is no throttle control or info on the monitor. Wiper and washer works fine from the controller, so there is coming power to the cabin.
I don't think you can take measurements on a CAN network using a multimeter because a multimeter only measures averages not the peaks and troughs of the signal. In truth it requires an oscilloscope so that you can see whether or not the data waveform on the network is a "clean" signal.
I understand that Nige. But, as far as I know, CAN lines have some voltages and resistances that can be measures with a multimeter. Not to know if there is info on the line, more like to know if there is any cable damaged.
I would not agree that a resistance test is a guarantee that a wire is in good condition. I would suggest that if you want to check continuity on the CAN lines (or any other wire in a harness for that matter) then the best method of doing so is to unplug the harness at both ends of a particular wire. Jumper one end of the wire to machine frame ground and supply battery+ voltage to the other end of the wire via a fused 5A test light. Many times a damaged wire can have continuity but is damaged to such an extent that it cannot either pass the appropriate current or a clean signal in the case of a CAN. I follow a few diagnosticians online, and I can tell you without fear of contradiction that it is entirely possible to measure what would be considered as "acceptable" voltages on CAN wires but that the signal quality is total garbage.
Thank you for that advice. I will try that. Also, something that I didn´t mention. Before the light on the machine ECM was flashing red/yellow. Now is stable green. I understand that means the ECM is ok.
So you have ET and it will not communicate with the machine at at all, correct? That narrows it down a little. You machine has 3 seperate independent data links: CDL, CAN1 and CAN2. As you can see from the diagram below, the Machine CAN on 252Green and 262Yellow is the Global CAN and is what communicates with ET. The other CAN is what's known as a local CAN. You are partly correct in testing the resistance of the network. Measuring the resistance across pins F&G of the ET connector should give around 60 Ohms with the battery disconnected - that's important. As soon as the battery is connected, modules may start to try to communicate on the network and make readings meaningless. This only gives a quick and dirty indication that the physical wiring and 120 Ohm resistors appear to be OK though. Nige is correct that measuring voltages of data wires is almost useless. It does however tell us that they are at least connected to something and reinforces the sesistance test. Yes, you have a short to ground somewhere. There should be open circuit between CAN wires and ground, with battery disconnected. Rather than trying to look for a physical short in the wiring, it's easier to look at what other modules are on the Global CAN, disconnect them one at a time and see if the short disappears. As you can see from the diagram, this means disconnecting the Monitor, Product link ECM, Engine ECM, Machine ECM, Keyreader module and gateway module one at a time. Note that your machine may not have all of the modules installed.
Cmark has nailed it. Only thing I’ll add is to check every single connector that carries the global CAN wires. Pull apart, check for moisture & corrosion, reconnect. Found this on a few NextGens.
Thank you Cmark and Chris. I measured the resistances with the battery connected, so I will measure again everything with the battery disconnected to test if it is ok.
Just wanted to say that your explanation was excellent and even logical. Thanks Simon C
Thanks. Just doing my job.
Hi Cmark, I tried yesterday to measure the resistance with the battery disconnected. I have no resistance (0L) with the battery disconnected. So maybe there is no ground somewhere. I really don´t know what else to test or check. Everything looks ok, cables, connectors. There is still no info in monitor, also not able to connect via ET
I have some updates about the machine. I found on the troubleshooting of the machine a 10-step guide about CAN issues. Everything is ok, except when measuring the voltage of the both of the CAN lines. With the key on and the battery connected, the voltage expected is 0.5v and 4.5v in each cable. I have, on one CAN line both around 4.5v and on the other line both 2.5v. I continue with the following steps, but they give the correct result. I kind of don´t know where to keep looking for the issue.
I don't quite understand what you mean. Are you testing both the Global CAN and Local CAN? Both wires of one twisted-pair are at 4.5v and both wires of the other twisted-pair are at 2.5v? Where are you taking the measurements? We are really only interested in the Global CAN measured at the ET connector. Can you show a screenshot of ET when it says it can't connect?
Is the attached the procedure you are referring to.? I think the text in Step 3 on Page 5 may be relevant.
Yes, I am testing both CAN lines. I am taking the measures at the ET connector. I will take a picture next time, but there is no connection. Yes, that´s the procedure I was doing. Point 3 was the one I referred before. Edit: There is something "new". The voltage of one of the can line (sorry I don´t remember which one) is around 4.5v, but when I disconnect the switch panel it goes to 0v. But still with that disconnected the machine is not working
Another update: I tested the voltage of the CAN on another Cat 320 that I have (older), and the voltage is around 3.5v, as stated in step 3 of the procedure that Nige sent. I will now take all the cabin wiring apart, as it was repaired several times and I do not trust so much in that repairing.
At the ET connector there is only one CAN line: The Global CAN. The other twisted pair is CDL (Cat Data Link). CDL runs a different protocol to CAN and you will get different voltage readings. Did you unplug all the modules on the Global can as suggested earlier? If so then plug them back in one at a time and see what voltage you get at the ET connector and also if ET will connect.
These are an approximation of how CDL and CAN J1939 work. As you can see, you would expect to see different voltages on CAN high vs CAN low. On CDL you would expect to see roughly the same voltages.
I don’t believe you have proved an ECM failure. The fact that bypassing damaged cab connectors restored starting, combined with both ECMs disappearing from ET, points toward a common power/ground or Global CAN problem. OL from either CAN wire to ground with the batteries disconnected is correct. Concentrate only on pins F and G at the service connector—252 Green and 262 Yellow—not the CDL pair. Disconnect each Global CAN module individually and retry ET. Also load-test every ECM supply and ground because seeing 24 volts on a multimeter can be misleading through a high-resistance connection. Don’t replace or reflash an ECM until the network wiring and module isolation tests identify it. My leading suspect is still the previously repaired cab harness or one of its bulkhead connectors. The fact that manipulating those connectors changed the machine from no-crank to running is more persuasive than the ECM LED or the averaged CAN-voltage readings. But what do I know…
Update: I unplug all the modules on the Global CAN and plug them one by one with ET connected. With Machine ECM only: error to connect ET. With Engine ECM only: error to connect ET. Both Machine ECM and Engine ECM: error to connect ET. Machine ECM and monitor: connect to ET, only monitor info. There is a connector from the ET to the wiring harness that goes outside the cabin to both ECM, y clean that one and still the same. I review all the connections that were bypassed and seems ok, none of them are CAN lines, they are all fuse and relay outputs. I did a load-test on every connection and connector. Everything looks fine.
You have not mentioned the Product link. Have you unplugged it as well?
Yes, that was unplugged also. I connected at the end but it was the same.
Something else that I didn´t mention yesterday, I also measure voltages of the Global CAN in the ET connector with everything disconnected and connecting one by one. The only one that gives voltage above 4.5 is connecting only Engine ECM, I have around 4.9v in the Global CAN. With Engine ECM and product link for example, I have around 2.5 on one cable and 1.7 in the other one. The same with everything else: Engine ECM, monitor, and dial.