Air conditioning has been around for a long time. I have been working on heavy-duty air conditioning systems for quite some time myself and to be honest, one certain aspect of these systems has changed dramatically — the electrical controls to the system. When I first started in this field 20-plus years ago, the controls were not as electronically savvy as they are now. I have seen quite a few changes over the years and in my opinion, sometimes these changes do not help as much as we think they do. I have seen the simple systems with just wire, fuse, thermostat, switches and a clutch to having a system with pressure transducers, stepper motors, and an electronically controlled thermostat. So with this new savvy technology, of course, the general public and some techs think that these systems are easier to diagnose because of the software that can communicate with the computer and give you data.
Well, I have news for you! On today’s modern electronic systems, if you don't have an oscilloscope and a good understanding of electronics and electrical fundamentals, you are going to be lost. Not only that you need to understand how the system works before you start testing. In this case study, I am going to show you why a DSO is needed and why it is so important to understand system operation and proper specifications.
A bus with a 1606 — clutch overload
This bus came in with a complaint of erratic air conditioning operation and is equipped with a TK3 System on it. I hooked up the laptop to check codes and found a Clutch Output Short code as indicated on the TK3 Clutch Output Short Picture. The code set criteria states that if the current exceeds 3.75 amps, there will be a 3-second delay before the alarm and if it goes over 6 amps there will be no delay in the alarm. So after seeing this, I look at the corrective action box on the troubleshooting manual and it gives you the basic things to check. Wiring, clutch and also to verify the current measured at the clutch reads the same as it does at the computer.
I have done all these checks and one spec does stick out as an alarm to me at that time. My Clutch Current looks too high to me on the software. I hook up my DSO (Digital Storage Oscilloscope) to the Clutch power and ground, and put a current clamp around the feed wire (Figure 1).
I turn on the A/C and have my scope set up for a single trigger (so I don’t have to worry about missing any data as I am at the front of the bus almost 40 feet away). So as you can see by my first markup I have good Battery Voltage, Good Ground, but the current is showing excessive on the scope as well as the software (Figure 2). Also, I looked at my Math channel shown in my next scope picture and the resistance showed it being lower than what I was expecting, which was around 11 to 12 ohms (Figure 3). After this, I got out my ohm meter to test the coil resistance and the resistance on the meter showed close to the same as it did on the Pico Math Channel. I am convinced the clutch coil is shorted.
Before we move on to the rest of the story, did you notice on the clutch current that it had a dip midway up the ramp and then went back to normal? What you are seeing is the mechanical movement of the clutch. How can you see mechanical movement electrically? This is a very good question and it made me ask the same question myself.
I decided to ask my buddies at Autonerdz why this happens. It all has to do with Lenz Law. Now I am not going to go deep into the theory on this but after talking to quite a few Autonerdz members, Ian Crane, Spencer Decordre, and the one and only Autonerdz team of Tom Roberts, Brian, and Sean it all comes down to this — what you are seeing is magnetic fields that are working against each other at the moment of armature movement. Since the armature moves as the clutch is applied it momentarily changes the direction of current flow because of the magnetic fields working against each other between the clutch coil and the armature.
Moving on now
Alright so now back to the problem at hand. I went to my parts guy, ordered a new coil and that's it, right? Nope, if anyone has met me they know that a lot of the stuff I share with everyone is not always cut and dry. Here is the rest of the story. I went on vacation for a week and when I came back I heard the new coil was installed but the bus was back once again for intermittent A/C.
I pretty much felt like the Koala Bear from all those Facebook posts showing the Koala with its mouth hanging out with script on the picture saying, "The look on your face when a vehicle you worked on previously is back in the stall." Face it, everyone, I'm human just like everybody else and I make mistakes. However, I learn from my mistakes. So with that being said, I decided to check the resistance of this new coil and it was the same. I then started looking at the service manuals and I pulled out one for a 2011 model and one for a 2016 model. Well due to my laziness here, I’ll admit it, I neglected to notice in the service manual that the coil specs had changed from 2011 to 2016. The bus I’m working on has a resistance spec of 7.1 ohms at 100 degrees F and at the lowest temp of 50 degrees F, it is at 6.6 ohms. On the older models for that specific compressor and coil setup the resistance is 12 ohms at 100 degrees F and 11.2 ohms at 50 Degrees F. So I am an ohm off here but the new one matched the old one. So, I am crossing this off the list as a problem.
After this, I hook up to check codes and it does not have any codes. Oh great! After seeing this I am drawing on some prior experience with this system, to be honest, to isolate this issue. I hope this information will help everyone out from having the headaches I had at first when I first started working on these systems. I decided to go to the Electronic EPR Valve.
I bet you're asking Mike what is an EEPR Valve? What this valve does is it controls the backpressure in the evaporator for more efficient cooling. So if this valve is malfunctioning it can cause the backpressure in the evaporator to either go low or high causing intermittent cooling problems.
The troubleshooting tree for this is nonexistent. All that the Thermo King manuals tell me is that I that it is a two-phase stepper motor that uses one phase to go one direction and the other phase in the opposite direction. The operation of this valve is a little vague, to be honest, but by using the scope we can have a better idea of how this controller actually turns on the valve. So, to see if I have any activity on the EEPR Valve I take two low amp clamps and put one clamp on each leg and use the TK Software to make the valve turn on and off. There is no activity whatsoever.
I then checked the output voltage of the valve at the computer on all four wires (Figure 4) and I had no voltage. After this, I checked all powers and grounds to the module and they passed. So what has happened here is the computer has failed. I then proceeded to check the EEPR Valve for shorts with a DVOM thinking the valve took it out but I saw no shorts to ground on the valve. I cannot give a spec unfortunately because the TK Manual does not provide a resistance spec for the Valve. So at this point after verifying the wires to the valve were not shorted to ground I was concluding that the board just failed open.
Another win for the Pico!
I ordered a new computer for this unit and got it installed the next week. Before I connected the EEPR Valve though I left the Valve disconnected to see how all the legs worked with the circuit open. All legs show it being Pulsed with a Positive Pulse Width Modulation at 28 Volts. Also, the computer did not log a code while I did the test. So for this Controller, it has not been programmed into the software to notice a fault. The only way you are going to know to go this route is with prior experience, or you the reader just learning it from me. Now here is where the fun begins and yet proves beyond a shadow of a doubt that a scope is needed now in the HD Industry to fix these problems. I hooked up both current clamps with one clamp on each leg again and then also looked at a power and ground on one leg as well. As you can see the current is spiking over 3 amps (Figure 5)! That will definitely take out a Computer for this system.
I could not see this issue with the DVOM because that is a static test and the valve is cold. After the circuit heats up you can see it fail and with the scope, you can catch it in the act. Also look at the period of time the fault happens, 76.89 ms. Good luck ladies and gents trying to see that on a DVOM. Even with Min/Max, your cutting it close. So what took out the computer was the valve. It was intermittently shorting to ground internally and damaged the computer. I then installed a new valve and saved the new computer and then did another capture. On the good capture here displayed I am showing 246 milliamps peak to peak on this valve when I cycle it with the computer and I have the 24 Volt Pulse Width and a ground showing voltage as well but it is different. On my ground I am seeing the voltage rise on the turn-on of the valve and then when it turns off the voltage tapers off. Also notice that every time Voltage is taken away or applied it shows a spike in voltage (Figure 6). This has to do with the magnetic fields of both field coils working against each other plus after making some more measurements of the harness I found 150 ohms on each ground leg. So the resistors are being used as current limiting devices for the valve. That explains the elevated ground voltage. On my known good valve, I have 468 milliamps. This vehicle is fixed. I hope everyone enjoyed that. It just goes to show that the software does not tell you everything and it can lie to you. Using a scope now is more important than ever.
