Wohlenberg Guillotine Multiple Faults

We were called out to look at a Wohlenberg MCS-2-TV guillotine for a local packaging company. The machine’s video display was gradually fading out to a blank screen (video demonstration below). The display would come back after turning the machine off for a few minutes, only to fade out again. The first thing we noticed is that battery acid had leaked onto the CPU board due to overcharging of the battery. Being very cautious not to lose the battery-backed parameters of the machine we replaced the battery and found that no tracks were damaged on the CPU board. The next thing we did was check the power supplies – we saw very little ripple and no dropping out. This led us on to video display board. We found the video data signal had less and less information as the video faded out. We concluded that the fault was coming from one of the circuit boards in the guillotine. We pulled the electronics out of the machine and took it back to the workshop so that we could more quickly find and fix the fault.

After installing the electronics in the workshop and copying down the machine parameters, we first checked with a thermal imaging camera whether any components were overheating. One of the processor chips was going over its maximum operating temperature of 75°C. Putting a finger on the chip dissipated some of the heat and the video display came back to life! After doing some circuit tracing we determined that one processor handled digital I/O and the other (hot) processor handled video output, machine parameter storage and keyboard interface. The processors had programs stored in EPROM chips and were able to communicate with each other via a RAM chip and octal buffer chips. We performed a reverse diode test on the address and data bus of each of the processors and found some low readings on one of the data pins. We then performed this test on each of the chips connected to the data bus – the previously mentioned octal buffers were the cause of the low readings. After replacing the octal buffers the video stopped fading out.

reverse diode testing

Results from reverse diode testing – it can be seen that data bus pin 7 tested lower than the others.

We installed the electronics back inside the guillotine, tested that it could cut properly and that the safety mechanisms were functioning and thought that was the end of the problem. However, the very next day the customer called us saying that the operators could not create or execute any of their cutting programs. This was severely limiting their production because they use the guillotine in automatic mode whereby the guillotine cuts then pushes the material forward, makes another cut, and so on. After reading the user manual and following the program creation steps, we confirmed that the user program screen could not be accessed. We found this to be strange considering that we could access and navigate through the main screen and all of the settings screens. We took the electronics back to the workshop. We copied the processor programs stored on the EPROM chips and traced out the circuitry related to address decoding in order to understand how the programs worked so that we could modify them and fix the fault. However, when we were up to the stage of being able to display our company name within the program we were able to obtain the machine service manual from Wohlenberg. The service manual allowed us to run a memory test, which confirmed one of the EPROM memory chips was corrupted. Luckily one of our other customers had the same machine and allowed us to copy the program from their EPROM chip and burn it onto a new chip. This fixed the fault of not being able to access the user program screen.

memory test

Memory test results showing a corrupted EPROM.

We installed the electronics back in the machine and found we could create and execute cutting programs. However, the next day the customer called us and said the guillotine was still not fixed! The machine had worked for several hours but then it stopped and the display turned off and the guillotine had to be restarted. However, every time it was restarted it would work for a few minutes and stop again. After examining the input and output LEDs on the CPU board we found that the “On” signal was turning off when the machine stopped working. We previously found in the workshop that this input had to continuously be on, otherwise the display would go blank and the machine would beep, which was happening here. After studying the relay logic in the schematics, we found that there were many things that could turn the “On” signal off, such as a knife CAM overrun or the light barriers not working properly. We systematically worked through these layers of logic using relays with LED indicators and found that a microswitch for the safety bolt was burnt-out. After replacing this microswitch the customer has been running the guillotine without issue.

image4-1

Relay board (left) and safety bolt with microswitch (right).