Author Topic: PSU Overshoot: Not really repair - ReDesign.  (Read 6206 times)

0 Members and 1 Guest are viewing this topic.

Offline janengelbrecht

  • Regular Contributor
  • *
  • Posts: 180
  • Country: dk
    • JP-Electronics
PSU Overshoot: Not really repair - ReDesign.
« on: November 15, 2014, 09:23:45 pm »
i have a BaseTech BT-305 Power Supply : 0-30V / 0-3A. The problem is it overshoots when you press the OUTPUT button.
Schematics is here: http://janengelbrechtpedersen.dk/psu.PNG
And here http://janengelbrechtpedersen.dk/psu2.PNG

The OUTPUT control button seems to control the relay with contact set: K1a and K2 ? Or is it only K2 ? hmm.

Now I sit here and try to solve the issue in an easy way....cant live with it..the 1.5-2V overshoot can kill equipment like Microcontrollers etc.

(I know its a crappy PSU that Conrad sells :P And i have told them on the website...still they gave me 50 dkr in bonus for the review :P : Dont BUY this piece of shit: http://www.conradelektronik.dk/%7c-Regulerbar-laboratoriestr%f8mforsyning,-regulerbar-laboratoriestr%f8mforsyning-%7c150-W-0-30-V%2fDC-%2f-0-5-A.htm?websale8=conrad-dk&pi=513812&ci=SHOP_AREA_17452_2110330 )
« Last Edit: November 15, 2014, 09:37:00 pm by janengelbrecht »
 

Offline janengelbrecht

  • Regular Contributor
  • *
  • Posts: 180
  • Country: dk
    • JP-Electronics
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #1 on: November 15, 2014, 09:46:44 pm »
Ideas anyone ? What to do ? :)

Online wraper

  • Supporter
  • ****
  • Posts: 11509
  • Country: lv
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #2 on: November 15, 2014, 11:13:34 pm »
Seems that you get overshoot when K2 relay switches and C16/C17 get charged. When PSU is recovering output voltage, you get overshoot. I think you could try to increase C3 or R7 value, however you will slow down transient response. But as you might get overshoots on transients too, it might be not too bad actually.
« Last Edit: November 16, 2014, 01:47:32 pm by wraper »
 

Online wraper

  • Supporter
  • ****
  • Posts: 11509
  • Country: lv
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #3 on: November 15, 2014, 11:16:26 pm »
Could also try to decrease C5 value. But of course should check transient response before/after mod.
« Last Edit: November 16, 2014, 01:47:46 pm by wraper »
 

Online wraper

  • Supporter
  • ****
  • Posts: 11509
  • Country: lv
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #4 on: November 16, 2014, 01:47:17 pm »
Actually looking at the schematic one more time, seems that you should do completely opposite thing to what I recommended previously. IMO, the problem cause is how current limiting is made. Voltage and current control loops basically are independent of each other but are connected to the same power stage input. Wins the loop which have lower voltage on the output because of the V3 and V4. When CC loop wins (charging capacitors), CV loop output voltage shoots up to the max but don't affect output voltage yet because of the V3. Once current drops below the limit, CC loop releases the output, but as CV loop output is on the max at the moment, here you go - you have overshoot as CV loop cannot drop down voltage before V3 fast enough.
« Last Edit: November 16, 2014, 02:02:22 pm by wraper »
 

Online wraper

  • Supporter
  • ****
  • Posts: 11509
  • Country: lv
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #5 on: November 16, 2014, 02:17:42 pm »
Imo you could make the CC control loop rather than working directly with power stage, decrease CV loop reference voltage somewhere around R6 instead. Seems that one inverting opamp (connectod instead of V4) + diode on opamp output could do the job. The trick is to keep this thing stable in all conditions which might be not so trivial.
 

Offline Kevin.D

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: england
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #6 on: November 16, 2014, 04:05:16 pm »
Hi there .
Just posting that 'my power power supply overshoots when I press the output' isn't enough information really to say  what could be causing it, It could be 3 or 4 things.
You need to do some tests to establish the cause before you or anyone else can suggest a fix. Like
Does  the overshoot prob dissapear if you short out R38 ?(which might suggest a relay syncro/contact resistance  cause).
Or does the supply also overshoot when  switching from CC to CV ? , this though would be  unusual due to the large output caps (which have to recharge and thus prevent overshoot whenever entering CV from CC  mode). test with  Fet/BJT and a series resistor to limit current to say .5A to act as a load stepper, with this on the power supply output and the supply current limit set below .5A drive the Fet with a pulse from a sig gen a couple of cycles so that the supply  alternates between CC -CV mode and back then see if there is any overshoot on your scope. 
 Then use the same simple step load to step only the CV mode between two load resistance values (just add another resistor in parrallel with the load stepper), but this time turn up your Current cntrl to full so the supply remains in CV mode during switching and observe if your CV loop is overshooting.
These 3 tests should give you a pointer to where the problem is.

Regards   

p.s You say this is a 'crappy supply so dont buy to ppl' but looking at the schematic it looks alot better than many others I have seen, Is this a rebadged mastech ?. They haven't skimped alot on the main circuit at least .
« Last Edit: November 16, 2014, 09:46:11 pm by Kevin.D »
 

Offline janengelbrecht

  • Regular Contributor
  • *
  • Posts: 180
  • Country: dk
    • JP-Electronics
Re: PSU Overshoot: Not really repair - ReDesign.
« Reply #7 on: November 17, 2014, 02:30:34 am »
Hi there .
Just posting that 'my power power supply overshoots when I press the output' isn't enough information really to say  what could be causing it, It could be 3 or 4 things.
You need to do some tests to establish the cause before you or anyone else can suggest a fix. Like
Does  the overshoot prob dissapear if you short out R38 ?(which might suggest a relay syncro/contact resistance  cause).
Or does the supply also overshoot when  switching from CC to CV ? , this though would be  unusual due to the large output caps (which have to recharge and thus prevent overshoot whenever entering CV from CC  mode). test with  Fet/BJT and a series resistor to limit current to say .5A to act as a load stepper, with this on the power supply output and the supply current limit set below .5A drive the Fet with a pulse from a sig gen a couple of cycles so that the supply  alternates between CC -CV mode and back then see if there is any overshoot on your scope. 
 Then use the same simple step load to step only the CV mode between two load resistance values (just add another resistor in parrallel with the load stepper), but this time turn up your Current cntrl to full so the supply remains in CV mode during switching and observe if your CV loop is overshooting.
These 3 tests should give you a pointer to where the problem is.

Regards   

p.s You say this is a 'crappy supply so dont buy to ppl' but looking at the schematic it looks alot better than many others I have seen, Is this a rebadged mastech ?. They haven't skimped alot on the main circuit at least .

Thx for your inputs...I will make the measurements next week and return here with a follow up.
I have checked around on the internet...its a problem all owners of this power supply has...so a redesign is a must.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf