Steelseries Arctis 7 headset in Linux

Repairing an Oltronix B602D Power Supply

 This is a repair/maintenance description of a 70's linear bench power supply. Oltronix LABPAC B602D.

I bought this old power supply from a scrap dealer. Haggled a bit and got a good price. It's a nice 70's linear laboratory power supply with two channels, 0-30V 1.4A and 0-60V 0.7A. The two channels are fully independent with isolation voltage between them and chassis +-500V DC (according to the manual). And it has a "digital" voltage display! Oltronix was a company that was based in Sweden that made high end laboratory equipment. Parts of the company still exist and are based in the Netherlands. 

There is a similar three channel model B603D that has an additional 6V power supply.

 

 I opened it an cleaned some dust, before powering it on. Take it apart (as Dave Jones says)! It didn't look too bad. So I then powered it up. The 0-30V channel worked fine. But the 0-60V channel did not. So I began some more detailed tear-down and testing.

 

Repair

Things that I observed and fixed:

- The transformer can be selected for 220V or 240V with a soldered jumper. I changed it to 240V, because nowadays we have practically 240V (nominally 230V). 

- Took out output transistors (Darlington NPN 2N6385). Tested them. They tested fine, except one that showed only half the hFE of the others. However, it's in the crude "pre-regulator" that is used to halve the voltage to the output transistor on lower voltages. So no problem, it works fine. Applied new heat sink paste and remounted.

- Pin headers for the separate regulator PCB and connectors for potentiometers were corroded. By mechanically cleaning them, I got the 0-60V channel working, too. But the pins looked so corroded, so I carefully desoldered them and put in new pin headers. 

- Current limit potentiometers. They were in bad shape and current limit was irregular. However, they were 500 ohm and 6.3mm shaft, so I had a hard time finding suitable replacement. Finally ordered a pair of Vishay 14910F0GJSX10501KA. The axle is a bit longer, but could potentially (sic) be cut down.

- Voltage adjustment potentiometers are high quality multi turn, mounted on an axle connected to the counter. They seem to work fine and could be a pain to renew, so hopefully they stay OK.

- The "digital" voltage display had some numbers missing, with paint flaking off from the two leftmost digit wheels that are made of some rubber material. I just painted new digits with white paint, but it doesn't look so good.

- User guide with schematics is available on the Internet. There is a version A and B schematic. It turned out this power supply is Version A. But I could find some issues with the schematics. Version B clearly has a short in opamp A30, input 5 and 6 (I corrected this in the schematic in the pictures shown below, but the pdf is unchanged)! Also diodes CR91-CR92 in this power supply are connected like in schematic Version B on the PCB, even though this is version A.

 

 

- The regulator circuit is mounted on a separate small PCB that can be removed. The boards are identical for the 30V and 60V channel. In the schematic, it's in the area enclosed in the dashed line. There is a 10 pin header that connects to the main PCB. It's a simple circuit with a 1N823 zener diode as reference and five opamps taking care of voltage, current limit, driving the output transistor and a separate opamp for driving voltage or current limit LED.

- Ammeter didn't work on the 60V channel, which turned out to be bad connection in the Volt/Amp switch and really corroded adjustment trimmer for the meter. I opted for desoldering the meter trimmers and getting new ones. I just lubed the switch, but if it doesn't work good in the future, I might have to replace it.

- I didn't replace any capacitors, because they tested and looked fine.

Schematics and user guide

I've uploaded a cleaned version of the pdf user guide at https://johanh.net/Oltronix/Oltronix_b602d_b603d_laboratory_supply_1977.pdf (15.5 MB). Note that the schematics contain some errors, as mentioned above. This isn't corrected in the pdf, only in the pictures above!

Observations about the design

- There is no voltage regulation for the power supply regulator part. It's about 32V (- to +).

- It's a floating design, so identical regulator circuits are used for the 30V and 60V channel (I checked the PCBs and they are identical; you can swap them).

- It has a temperature compensated zener diode 1N823 as reference voltage.

- In version B of the schematic they dropped the CR60 ZF12 zener diode and changed to a reference zener 1N4576 that is optimized for lower current. Some of Oltronix later 70's PSU series have the same 1N4576 zener. Maybe this was a cheaper alternative. In their 1984 703 series PSUs, they have replaced the zener with a TL431 shunt reference that is common component still today. Today, such temperature compensated zeners are expensive, whereas 431 compatible shunt references are very cheap.

Regulator PCB reverse engineering

I reverse engineered the regulator PCB, to make sure it matched the schematic. As I already mentioned, it was same as the Version A schematic, except for the two diodes CR91-CR92 that were connected like in the Version B schematic.

I created a copy of it in KiCAD, if I ever need to create a new PCB (or as a base for modifications). The KiCAD project is available here: https://github.com/hevanaa/Oltronix_B602D_Power_Supply