Repair of KS94 Temperature Controller manufactured by PMA

I recently repaired a KS94 Temperature Controller manufactured by PMA. The power supply in the KS94 had failed. The Switch-Mode-Power-Supply's Switching Transistor was shorted. Because the transistor was shorted, I removed the Switching Transformer to make sure it was still good. With my DMM set to resistance mode, I found that the primary winding was shorted to another winding. I drew a schematic of the SMPS to determine if this was a correct function of the Switching Transformer. It was not. The primary winding shorted to the secondary winding. This Switching Transformer was custom manufactured for this application and was not available for purchase. The Switching Transformer was a sealed epoxy unit, so rewinding it would be near impossible without destroying it in the process of disassembly.

The owner of the KS94 Temperature Controller needed it back quickly for an urgent production run, and a new KS94 unit was weeks away if purchased. Thus, another solution had to be worked out.

Ron Abts had me work out the necessary voltage requirements for the KS94 Temperature Controller. Then he went to our parts department and pulled two power supplies off the shelf. The one power supply would produce the proper voltages for the CPU Logic Circuits and the Analog Circuits. The second power supply would make the required isolated voltages for the 4-20mA output. I installed the power supplies, and tested the KS94 Temperature Controller good. The PT-100 Temperature Sensor Circuit worked and the 4-20mA output tracked the Temperature Sensor's changes good.

There are three output relays in this KS94 Temperature Controller. I tested them as part of the KS94 tests. I found them to have high contact resistance, so I replaced them as well.

The KS94 Temperature Controller is in the customer's machine and working.

Repair of AF-300E$ General Electric / Fuji Electric Inverter Drive

A company in New Mexico sent us a Switch-Mode-Power-Supply/IGBT Firing Board from a General Electric / Fuji Electric AF-300E$ Inverter Drive. This AF-300E$ runs a 200 Horse Power Motor. The company only sent the boards in for repair because the drive was to too large to ship to us.

The part number of the Switch-Mode-Power-Supply / IGBT Firing Board is EP-3603A-C2-Z4. The Switch-Mode-Power-Supply Board had a blown trace and external smoke residue covering the board's conformal coating. I cleaned the board and drew a schematic of the blown trace circuit and found that it was the ground path for the High Voltage Bus Monitoring Circuit.

I then tested all the components on both the Power Supply Board and the IGBT Firing Board with my DMM. All the IGBT Pre-driver Transistors on the IGBT Firing Board were shorted.

Judging from the smoke residue and the shorted IGBT Pre-driver stages, we concluded that most likely the IGBT's in the drive were also shorted. I made the repairs to the SMPS Board and the IGBT Firing Board. I drew a schematic of the IGBT Firing Circuit to find out how they were being fired. Then powered up the board with 460VAC. I then tested the IGBT Firing Circuit and the Bus Monitoring Circuit with input signals. Each circuit functioned correctly. In addition to testing these two circuits, I tested good the voltages of the SMPS Circuit. These tests had to be done before mating the boards to the drive. Otherwise the drive would be blown again.

Because we knew that there had to be damage to the drive itself, Ron Abts flew out to the company in New Mexico to check out the drive. He found that the IGBT's were in fact shorted and a 600 Amp Bus Fuse was open. He also found that there was a problem with the Three-Phase-Bridge-Rectifier Circuit. This circuit was made up of four each parallel per line phase of Dual-Diode-Power Modules. Some of the diodes did not measure correctly with his DMM. He also found that the Main Contactor Pre-charge Resistor was open. He partially disassembled the drive to get the IGBT and Dual-Diode-Power Module part numbers. He phoned these part numbers back to the shop so that we could get an early jump on getting them in. He then recommended that the drive be shipped to us, so that the repairs to the drive could be completed. The company agreed and shipped us the drive.

Upon receiving the drive, I started the disassembly process to get down to the IGBT's so that I could replace them. This drive is huge. The current carrying members are large flat metal bars. And so I had to be very methodical in the disassembly of the drive, so that I could re-assemble it correctly. One part not re-installed or installed in error would mean the destruction of the drive again. And wiping out all the effort placed into the repair of the drive.

Once I got down to the IGBT's and Three Phase Bridge Power Modules, I was able to test them. The drive was using three parallel Dual-IGBT's per motor phase. The U phase IGBT's were not shorted, but failed to conduct on my IGBT Tester. The V and W phases were shorted through and through.

I tested the Main Contactor and found the contact resistance to be too high. I disassembled the Main Contactor and cleaned the contacts. I then replaced all the damaged power modules while Ron Abts and Charles Green worked the new pre-charge resistor into the drive. Once the drive was completely re-assembled, we ran the drive unloaded and viewed the IGBT Firing Signals and U,V,W Outputs with a Fluke Oscilloscope. All looked good. We then ran a motor.

Ron went back with the repaired drive to install it at the company site in New Mexico. The repaired drive is working in the machine.

SHIMA SEIKI SFG-III I/O PCB TESTER

There are many times in the electronic repair industry that the technician must be inventive to overcome an obstacle. For instance, I have worked on numerous SHIMA SEIKI S-252, SFG-III I/O-C PCB's. This I/O PCB routes the Input and Output Signals to the CPU-III PCB of the SHIMA SEIKI GLOVE KNITTING MACHINE. There are three M5M8255AP-5 Programmable Peripheral Interface IC's on the circuit board. The 8255's work for the CPU-III PCB in routing the machine I/O to and from the CPU-III's data bus. As the 8255's name states, they are programmable. So when the CPU-III PCB boots up, it sends the programming information to the I/O PCB's 8255's.

So being, the only way to test the correct function of the I/O PCB as a whole unit is to program the 8255's, and then command them to perform a function that is observable. In the past, I had to unsolder the 8255's and test them individually with an IC Tester. Then I would have to test the solenoid driver transistors, and the several support IC's with the diode test function of my digital multimeter. This was a very time consuming process, for the 8255's each had 40 pins that had to be unsoldered.

With this in mind, I invented a tester that connects to the SHIMA SEIKI I/O PCB. The tester programs the SHIMA SEIKI I/O PCB's 8255's for their correct I/O function. And once the three 8255's are programmed, I am able to test the solenoid driver transistor outputs, and the basic I/O functions of the entire I/O PCB. With my tester, I am able to zero in on the exact malfunction within seconds. Whereas, it used to take an hour of unsoldering and probing with a digital multimeter to find the cause of the malfunction. The time savings for the customer are enormous. Additionally there is no doubt that the SHIMA SEIKI I/O PCB will function properly in the machine after the the repairs are accomplished.

My SHIMA SEIKI I/O PCB TESTER is based upon the PIC16F84 Microcontroller, manufactured by Microchip Technology Inc. I programmed the PIC16F84 to first program the 8255's on the I/O PCB for their correct I/O function. Then through push buttons, the PIC16F84 performs the proper SHIMA SEIKI I/O PCB Test. A group of LED's give indication of test status to the test operator.

The invention of the SHIMA SEIKI I/O PCB TESTER has saved our customers great amounts of time and money.

FUJI ELECTRIC Servo-Drive Model RYG.50HA-RP Switching Transformer Re-wind

I recently repaired a Fuji Electric Servo-Drive, Model RYG.50HA-RP. The servo-drive would not power up when the 220VAC, 3-phase line voltage was supplied. No LED's would illuminate on the front panel.

I dissassembled the servo-drive and found that a very small switching transformer on the IGBT FIRING PCB had scorched wires that had burnt through the yellow insulation tape. The transformer was about the size of a dime. This transformer was used to make the isolated power supplies for the three high side and the three common low side firing channels for the motor's IGBT POWER MODULE. Written on the transformer was a part number 66300V. But upon research, we found that the part was not available for sale, nor was it currently manufactured.

Our Lead Engineer, Ron Abts, unwound one winding of the transformer to determine the diameter of the wire and the number of windings and direction of turn. I then removed the other bad windings, also noting the number of windings and direction of turn. In all, there were three bad windings that had to be replaced.

I rewound the three bad windings and soldered the transformer to the IGBT FIRING PCB. I then removed the IGBT POWER MODULE, so that I wouldn't destroy it if the transformer wasn't wound correctly. I applied 220VAC, 3-phase to the line input of the servo-drive and the LED's on the front panel illuminated. I checked the control voltages, the 300VDC Bus Voltage, and the IGBT Isolated Power Supply Voltages good with my digital multimeter.

Then I re-installed the IGBT POWER MODULE. Re-assembled the servo-drive and powered up the complete servo-drive good. The transformer re-wind was successful.

Indramat Servo Drive TDM3.2-020-300-W0 Repair

An REA Technologies,Inc. Field Technician was called out to troubleshoot a Ward 50 Inch Flexo machine at a customer's site. The customer wanted the machine repaired before the sun came up on the next day. The Field Technician found that the Indramat Servo Drive (Model Number TDM3.2-020-300-W0) would not properly rotate the motor when the drive was enabled. The customer had a spare Indramat Servo Drive of the same model. But when the Field Technician installed this spare Servo Drive, the +24/+-15 Power Good LED did not illuminate. Indicating that there was a problem with the power supply. The first Servo Drive's Power Good LED illuminated, so the problem was not with the Indramat Power Supply Unit.

The field technician brought both Servo Drives back to our shop for repair. He also brought the Motor, so that we could test the Servo Drive after the repairs were made.

I tested the Transistor Power Module, of the first Servo Drive, with my DMM Diode Test Mode. I found that the "V" Output stage was open. The "U" and "W" stages were intact. Also the driver transistors for the "V" Firing Channels were shorted. This Servo Drive was going to take new parts and much time to repair, so I concentrated my efforts on the second Servo Drive.

The Transistor Power Module of the second Servo Drive was good. I didn't have an Indramat Power Supply Unit, but I knew from previously working on Indramat Servo Drives, that I could use external power supplies to make the +/-15VDC and +24VDC for the Servo Drive. I could also make the +300VDC Bus Voltage from a Variable transformer, a bridge rectifier module, and high voltage bus capacitors.

I turned on the external power supplies, and the +24/+-15 Power Good LED was not illuminated, just as the Field Technician saw out at the customer's site. I measured + and - 15VDC good on the Control PCB. But then I noticed that the ammeter on my +24VDC Power Supply was registering 1.5 amps of current draw. One of the functions of the +24VDC Input to the Indramat Servo Drive is to provide the voltage to a secondary power supply on the Power PCB. This secondary power supply in the Servo Drive creates the isolated firing channel voltages on the Power PCB.

I removed the Power PCB so that I could replace the electrolytic capacitors. After removing the capacitors from the Power PCB, I found that four out of six 1000 microfarad, 16V electrolytic capacitors were completely shorted. This explained the abnormally high current draw from my +24VDC power supply.

I replaced all the electrolytic capacitors on the Power PCB and re-assembled the Servo Drive. Afterwards, I powered up the Servo Drive and the +24V/+-15V Power Good LED illuminated. Good Good.

The Field Technician was unable to remove the Drive's Motor cable and Feedback Cable from the machine, so while I was troubleshooting and repairing the second Servo Drive, he was making the cables from wire and connector pins we had on bench stock.

He connected the Motor Power Cable and Feedback Cable to the Servo Drive and Motor. I turned on the Power Supplies to provide the + and - 15VDC and +24VDC. I then turned on the +300VDC Power Supply. I enabled the drive and it ran the Servo Motor good. With the speed potentiometer input to the Servo Drive, I could run the motor CW and CCW with full speed control.

The Field Technician brought the repaired Indramat Servo Drive and Motor back to the customer site for installation in the machine. The Servo Drive performed perfectly. The owner of the machine was very pleased we were able to quickly get it running again.

The customer also wants us to repair the first Servo Drive, so that he can have a good spare.

I was assigned to repair a Siemens 6SC130-0FE00 SERVO AMPLIFIER CARD. I tested all the power transistors with my DMM's Diode Test Mode. The power transistors did not test shorted or leaky. I then tested the diodes with my DMM. I found two leaky diodes,part number SB130 (labeled V123 and V622 on the PCB). I replaced the diodes to make the repair.

I then had to build a test set to power up the Servo Amplifier Card, and test its ability to run a motor. The main challenge with powering up the Servo Amplifier Card was with the switching power supply circuit. The Servo Amplifier Card only has the Switching Transformer whose primary winding is connected to ribbon connector X211. The switching circuitry to drive the primary winding is located on another card that I did not have.

The primary winding resistance read less than one ohm. This meant that if my switching circuit were to lock up in the full on mode, the chances were very high that the primary winding would be damaged. Thus great care had to be made in choosing a MOSFET that could handle the high voltage and current to safely switch the primary winding.

Once I had my switching circuit built, I had to experiment with the frequency and high voltage until I had good logic and analog voltages on the card. From the age of the card, I knew that the ballpark frequency would be around 15kHz to 25kHz. The best frequency for this card was 16.7kHz. My high voltage for the switching circuit was created from an external bridge recifier and two high voltage electrolytic capacitors wired in series. I used a VARIAC (Variable Transformer) connected to the AC input of the bridge rectifier to adjust the DC Voltage across the high voltage capacitors.

Now that I was able to power up the Servo Amplifier Card, the next step was to build a way to fire the power transistor outputs to the motor. The Servo Amplifier Card has six opto-coupled inputs at the ribbon connector X211, that turn on the the power transistor stages. Three inputs are used to fire the "High Side" power transistors, and three inputs are used to fire the three "Low Side" power transistors. The motor's rotation is dependent upon how these six inputs are sequenced.

I knew that I did not have to fire each power transistor output stage in this sequence. I only needed to see that a stage fired to know that it was good. So I connected six push button switches to the correct input pins of ribbon connector X211. The only caution being that I absolutely would not push at the same time the button for the "High Side" and "Low Side" power transistor stages of the same output. If I were to do so, the high current of the bus voltage connected to the P200 and M200 terminals would shoot through that power transistor stage and destroy it.

To make the motor bus voltage, I connected another bridge rectifier and two high voltage electrolytic capacitors wired in series to the terminals P200 and M200. Again, I used another VARIAC (Variable Transformer) connected to the AC input of the bridge rectifier to adjust the DC Voltage across the high voltage capacitors. I then pushed only one of the buttons to fire a single power transistor stage. I viewed the output change with an oscilloscope. I then performed the same test with the other five power transistor stages. All the six transistor power outputs turned on correctly.

Powers Process Controls "535 Process Controller" and Building Test Equipment

Here at REA Technologies,Inc., we are sometimes tasked to build test equipment in order to complete a repair.

Recently I was given an Powers Process Controls "535 Process Controller" used in the production of furniture. It was not certain whether the controller was at fault, or the problem was somewhere else in the machine. So a complete test of the 535 Process Controller was requested.

According to the machine prints supplied by the customer, One input labeled "DELIVERY" required a J-Type Thermal Couple input. I satisfied this input with a Thermal Couple we had in bench stock. Two other inputs were labeled +RSP/-RSP and +PLATTEN/-PLATTEN. These two inputs were 0 to 20mA inputs. To satisfy these two inputs, I designed and built a variable milli-amp source. The milli-amp source was controlled by a potentiometer, so that I could set the proper milli-amp current for the 535 Process Controller's inputs.

Without satisfying the three inputs, the 535 Process Controller would power up with RSP, PV1, and PV2 alarms. These alarms would stop the process controller from continuing and prevent further testing.

With the Thermal Couple and my milli-amp source, I was able to test all the Inputs as well as control the functions of the Outputs. The 535 Process Controller functioned normally. One of our field technicians went back with the unit and found the actual cause of the problem in the machine. He performed the necessary repairs and got the customer's machine running again.

FANUC I/O PCB, Type A20B-1001-0550/058

Recently I worked on a FANUC I/O PCB, Type A20B-1001-0550/058 from a company in Nevada. Upon initial inspection I found the 6.3AMP fuse (designated F2) was blown and the fuse holder was destroyed. Traces were blown off the pcb in the vicinity of a FANUC Hybrid Module Part Number NFZ2BC6WO (designated DVA04). And the NFZ2BC6WO Hybrid Module was scorched on one side. As I was going through the pcb in search of other problems, I also found an MB74LS240 IC (designated IC7) with a shorted pin 18. This pin drives one of the inputs to the Hybrid Module at position DVA04. In addition to this the Hybrid Module Input Enable circuit had two PNP Transistors that were shorted. This board was hit hard. Luckily, though, the control logic section was intact, as I was able to power up the pcb, and found life in this area. And no IC was drawing a large current and getting hot.

Before I fixed any of the found problems, I had to determine if the NFZ2BC6WO FANUC Hybrid Module was serviceable. I reverse engineered the Hybrid Module connections on the pcb to find the inputs and outputs. From this and consultation with our lead engineer, we determined that the outputs were TRIAC based. And the inputs were Opto-Coupled. I removed the suspect Hybrid Module, and set up a test circuit. The Hybrid module passed our test. It survived the explosion.

Inside the 6.3A fuse holder (F2) is a blown fuse switch. This switch was melted. One side of the switch goes to a red LED (designated LED3) that indicates a blown fuse when illuminated. The other side of the switch is tied to IC7's ground. What appears to have happened is that when the fuse blew (due to external circumstances), the Higher Voltage arced over to the fuse-blown switch and travelled into IC7 through the ground. The trace between the fuse-blown switch and IC7's ground was blown off the pcb.

I removed and replaced the shorted IC and Transistors. I repaired the blown traces and re-installed the Hybrid Module. And I replaced the fuse holder and installed a new fuse to fix the FANUC I/O PCB. I then tested the other TRIAC Hybrid Modules good as well as the Hybrid Module Input Enable circuit.

We shipped the FANUC I/O PCB back to the customer and when they received it they installed it in their machine. Afterwards, they gave us a call to let us know that it was working and they were very pleased.

Parker Compumotor ZETA6104 Indexer/Drive

Our lead field technician brought in a Parker Hannifin Corporation Compumotor ZETA6104 Indexer / Drive that the customer said had been functioning strangely. The ZETA6104 was used in a tool changer application. After disassembling the ZETA6104 and finding no obviously damaged components, I applied 120VAC to the line input. The control voltage power supply did not come up. It appears that the customer’s description of “functioning strangely” was due to the +5VDC power supply starting to fail. And when I tested the ZETA6104, the power supply had given up entirely.

After fixing the power supply, our lead engineer downloaded and saved the ZETA6104’s Program Initialization Data and Tool Changing Program to a laptop computer. This being done on the chance it could be lost in the future and require re-loading.

To test the ZETA6104 in this application, I built an interface board to simulate the machine I/O functions. Our lead engineer studied the tool changer program code to figure out how to properly sequence the inputs to simulate the machine. With the interface board I completely tested the ZETA6104 as if it were running in the machine. The ZETA6104 passed all the tests.

The field technician installed the ZETA6104 Indexer / Drive at the customer site and it performed without error.

Schleicher PROMODUL-U PLC Rack

A customer sent us a Schleicher PROMODUL-U PLC Rack from a Burkle Hot Press. The UNG 230 Power Supply was inoperable. We disassembled the power supply and found two burnt resistors in the Switch Mode Power Supply MOSFET circuit. The resistors had burnt open and the color code was charred. So determining the resistor value had to be found another way. I also found that the two MOSFET’s in the switching circuit were shorted. I reverse engineered (drew a schematic) of this circuit so that we could determine what the purpose and Ohmic value of the burnt resistors were. This would allow us to replace them with the proper resistance values. And in the process of reverse engineering, I tested the other components involved in the Switch Mode Power Supply circuit. I found that the Current Mode PWM IC was inoperable. I replaced the bad IC and shorted MOSFET’s. And through the reverse engineering process found that the burnt resistors were in the source to bus-ground connection of the MOSFET, and used for current sense feed back to the Current Mode PWM IC. From this knowledge we were able to replace the resistors with the correct values.

I performed the initial power up with the UNG 230 removed from the PROMODUL-U PLC Rack so that I could test the +5VDC and +24VDC output voltages at its edge connector. I gradually loaded the +5VDC output with an external resistor, eventually drawing 5 Amps from the UNG 230 Power Supply Unit. The +5VDC and +24VDC outputs maintained excellent regulation with no voltage ripple.

The UNG 230 Power Supply Unit performed excellently in the PROMODUL-U PLC Rack.