[SOLVED] Korad KD3005 with limited output voltage

[ElVectore]

Hello,

I have a Korad KD3005 power supply (0-30V, 0-5A) for about three years now. Since two years it is broken.

My power supply is the non-programmable version of the Korad or RND KA3005P power supply, from which I attached the schematics I've found in this forum.
Additionally, I added the calibration procedure for the KD3005 power supply, which is from this forum too.
Furthermore, there are photos from the problem attached.

But now to the problems:

• If turned on, the output voltage is 0V and the set-voltage is X5.35V (4-Digit-Display) while the set-current is X.535A. The X stands for a invalid character. To enable an output voltage, I have to go in the set-mode and rotate the first character to the left until the digits appear. Just like in this video https://youtu.be/ugdTeHeu-i0?t=42. If both set-values are in the valid range, it is unable to go over the maximum values again (30V, 5A).
• After the set-values are in their valid ranges, we come to the next problem. The displayed values are invalid. If I set the voltage to 30V the displayed value is for example 0.25V, and the real value is 17.5V. I created an table for the set, real, and displayed voltages:
  

  V_Set in mV	V_Real in mV	V_Display in mV	Q10_Volt	Q11_Volt	LEDs
  0	244	0	12	12	CV
  100	746
  500	2845
  1000	5438	0.05
  1500	8000
  2000	10620	0.13	12	12	CV
  2500	13220
  3000	15780	0.21	12	12	CV
  3100	16340		12	12	CV
  3200	16870		12	12	CV
  3300	17340		12	12	CV
  3400	17550		12	12	CC
  3500	17530		12	12	CC
  3600	17520	0.24	12	12	CC

  . I left out a few lines for readability, but you have to believe me that V_Real is proportional to V_Set until about it is 3.3V.
  
• The real-voltages obviously never gets over a voltage of about 18V. Furthermore, I never hear a clicking noise of the two relays on the power board which are driven by Q10 and Q11 on the main board. Regardless of which set-voltage is applied. If I bridge the voltage over the drivers, so that the voltage over them are set to 0V, the two relays are clicking. But this does not change the output voltage.
  

As a possible solution, I tried to get into the calibration procedure (V_Set == 12.34V and reboot), but it does not work. After each reboot the set-voltage and the set-current is set to 5.35V and 0.535A. Maybe the EEPROM is broken? To check I thought about building an I2C sniffer with an Arduino or a Raspberry, but I did not realize it yet.

Does someone of you know how to fix these problems?

[m k]

Maybe you've just lost one of those internal regulators.
Since it tops around half perhaps it's -12V.

[ElVectore]

Thank you for your reply m_k.
The internal regulators were fine.

After two further days of testing and checking almost every part, I've solved the problem.
The EEPROM was indeed the cause.
One or both of the I2C-Bus-Pins of the MCU were not soldered properly. A slight drag with the soldering iron over the pins solved the problem.
Both pins are marked in the attached photo. Please don't mind the poor looking EEPROM. I tried and successfully read out the data with a Raspberry Pi and python-smbus2. The pull-up resistor shown on the picture was not required.

As a bonus I added a simplified LTSpice Model of the power-board. Simplified in terms of not using exactly the used parts by Korad. Maybe someone can make use of it.

Edit: I just realized, that the marked Pins on the picture are the fan pin an the SCL pin. So not the first both pins are the I2C pins, but the second and the third pin (SDA).

[Mosaic]

Need a little advice.
My KD3005D has a failed 12V fan power feed.
Got around that by adding a 7812 TO220 Vreg to drive the fan, off the auxiliary xformer winding that feeds the PCB via the 3 pin  J3. Fan draws about 180mA.
However, my unit can only do 1.5A continuous (a 12V 35W automotive incandescent lamp used as a test load). The problem predates the fan fix.
Once it crosses 2.2A it will totally freak out and start click cycling the relays continuously on a 1 second time base or so with resultant current/voltage spiking.

Any suggestions where to approach this given I don't have a schematic, unless the one posted here is up to date for the KD3005D.  I have a decent workshop....Several O' scopes (Tektronix Analog & Digital up to 4GHz), Bench 6.5 digit DMMs , handheld Fluke DMMs, Spectrum analyzer, HP Vector NA, 3Ghz Agilent NF analyzer, Function gens, Capacitor & ESR testers, curve tracers...assorted other vintage Tektronix gear/plugins and 44 years of electronics experience and prod. dev. M.I.T. certified.

I could try to reverse engineer the control system for the 2 NPN power output transistors, but time is a bit tight. They are in full parallel configuration and this is not a switching PSU. Linear tracking with a beefy transformer running fed by 120VAC 60Hz.

So...since this thread seems the mot detailed in so far as circuit function...I am asking for some precision pointers. Thx!!