EEVblog Electronics Community Forum

Electronics => Repair => Topic started by: guido on February 25, 2018, 09:22:58 pm

Title: BK 8500 repair attempt
Post by: guido on February 25, 2018, 09:22:58 pm
Hello all,

Bought this load for about 30 euro's recently, but then obviously there is a catch.
So the IRF went bang, taking with it a big trace and some related parts. I took it out and i will replace also the current resistor, drive resistor etc. I already replaced the related TL084 to be sure. The M9711 schematic is a great help here (it's in one of the treads here). Some parts have a different value or are absent, but it is similar. Actually there is a separate current and voltage error amp, the output is switched with a relay towards the TL084s. Different from the simplified schematic from Gerry Sweeney.

Fixing the IRF related parts should be do-able. Same for the circuit that pulls the drive voltage towards the IRFs down (Q7/Q8 bottom right in the M9711 schematic). It is now pulling down all the time, so the regulation does not take place.

But the voltage monitoring is a bigger problem at the moment. With 10V in, i get 8.96 on the display. The circuit seems quite simple: A voltage divider and then straight to the ADC. R2 270k, R9 4k7 and i think a zener diode to protect the ADC input (D1). Goes to U2, pin 9. Measuring the voltage on the ADC input, it is correct (10V = 0.17V divided). So the ADC is damaged (guess not) or the calibration is way off. Would be strange if the calibration would be affected by an IRF short...?  There is a voltage reference IC present, this seems ok (ADR421, 2.5V). The voltage on the display tracks the input, it is just way off. Guess i can fix the error by tweaking the divider... But i don't know yet about the other ADC inputs used for AD VOUT (TPV) and AD IOUT (TPA). Need to fix the broken stuff first before i can test that.

So i looked at the serial interface next, since there are calibration commands present. Got an USB to serial adapter and hooked it up. Got the BK remote control program to work. Next was installing realterm on the old laptop and playing with the 26 byte hex commands:

6a return product model, serial number and firmware version gives a plausible answer
61 instrument calibration state returns 'protected'
68 read calibration info gives FF FF .... Assume it was not set by the factory or the eeprom is dead
60 01 enter instrument calibration state gives unrecognized command (same for 60 00, leave calibration state)
69 restore factory default calibration data gives unrecognized command

I assume the error unrecognized command is returned because the password used for 60 is incorrect. I tried both options in the manual; 8511 and 8512.
I also tried 8500, 0000 and ffff, still the same b0 response: unrecognized command.

Does anyone know how this password mechanism works on a BK 85xx load?  I assume it does not relate to the key lock password you can set.
For the M9711 i found a document that you can also enter the calibration stuff via the frontpanel (quickly hitting 4 keys in a sequence). Does not work with the BK.

One other remark is that the fans are constantly on at a modest speed. I'm wondering if that is ok. But i cannot test the dynamic behavior at the moment ;)

I anyone has info to share, please do. Thanks.
Title: Re: BK 8500 repair attempt
Post by: dacman on February 27, 2018, 11:24:10 pm
BK gave me the alignment software, but I had to ask for it.  (It is not shown on their website.)

I know of an 85XX unit that was powered off with a power supply attached (still supplying power I presume), and it cooked the BK.  I highly recommend removing DC power prior to turning off these units (or any electronic load).

The COM port uses TTL levels.  The BK serial adapters are isolated (and both USB and RS-232C units are available).  Does your serial adapter use TTL levels and is it isolated?
Title: Re: BK 8500 repair attempt
Post by: guido on February 28, 2018, 07:25:23 pm
Thanks, i will try them.

The serial adapter works fine with their remote control program and other commands. Just the calibration commands don't. It is a standard USB-serial 5V thingie. Not isolated, but i'm using an isolation transformer for the BK. My next action was sniffing the EEPROM, so see if the password would be one of the parameters. But i'll try to get the cal program first.
Title: Re: BK 8500 repair attempt
Post by: dacman on March 01, 2018, 02:40:13 am
Your serial adapter may be working fine, but here is a warning on page 46 of 76 of the manual:

"Do not connect the DC Load's DB9 connector to a standard RS-232
instrument. Doing so may damage the instrument, as the instrument
requires TTL logic signals, not standard RS-232 voltages."
Title: Re: BK 8500 repair attempt
Post by: guido on March 01, 2018, 11:10:48 pm

My serial adapter is a standard USB-serial 5V thingie. So 0-5V levels and not RS232. It works.

I got the program from BK  :-+
It needs NI VISA, so i had to intall that too  :--

Got the program working on a Vista laptop and calibrated the voltage part. Succes :)  The input voltage is measured correct now. Weird that somehow the calibration got lost.

So now i need to order the parts to fix the other problems. Then i can attempt a full cal.

And the fans. Again, they are both running at a modest speed at idle. I'm wondering if this is normal, or if there is also something wrong with the fan control circuit ??