Author Topic: MPJA Benchtop power supply, only 30mA-50mA short circuit current  (Read 7262 times)

0 Members and 1 Guest are viewing this topic.

Offline matt6ft9Topic starter

  • Contributor
  • Posts: 42
  • Country: us
I have two of the MPJA 14602PS bench power supplies.  ( maybe a re-branded Mastech)  They are a single output 50V, 3Amp linear benchtop power supply.  The power supplies are about 10 years old and are having an issue with the short circuit current.  The problem is very consistent; 30mA to 50mA is the max short circuit current throughout the voltage range.  (The constant current LED comes on when the leads are shorted)

First thing I checked was the “fine” and “coarse” pots for the current limit.  The fine pot is 1K and the coarse pot is a 6.8K pot.  Both seem to operate normally using a multimeter to check on them.  Next thing I checked was main cap.  It is a 4700uF, 100V Nippon Chemi-Con (fake?).  The cap was removed from the circuit and has an ESR of 0.8 ohms.  The cap value was read at 4300uF, which seems reasonable. 

One interesting item is if I put a 12V LED board on one of the power supplies, the current from the supply will be at the correct 150mA.  I don't have an electronic load to test where the power supplies go into constant current mode.

Attached are some photos.  On the front panel image, the supply on the left has its leads shorted out and it’s in constant current mode.  I have found a similar schematic, but does not seem be the correct version for this particular model.

Are there any ideas what would cause a maximum short circuit current of 30-50mA on two bench supplies?
 

Offline jeroen79

  • Frequent Contributor
  • **
  • Posts: 529
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #1 on: April 30, 2016, 03:37:33 am »
The scematic you posted wouldn't belong to a 50V/3A powersupply.
The large number of 2N3055's and the designation HY3020 suggest a 30V/20A supply, which the Mastech 3020 is.
The HY5003 would be a better match.

Assuming that the powersupply uses the same principles as the one in the scematic it will work like this:

The positive output is the ground.
Between the ground and output transistors there are resistors to sense the current, the end attached to the emitters will be raised above ground.
This voltage will be sent to the - input of an opamp.
A reference voltage is created above ground by the 12V regulator and the coarse and fine pots.
This is sent to the + input of the opamp.
The opamp can then turn off the output transistors by diverting the base current away from them.
This is basic opamp function: Sink current if - is above +.

Try to identify these parts and measure how they behave.
For example, is the reference voltage misbehaving?

Are both units behaving exactly the same?

Have you ruled out defective meters on the power supplies?

Instead of an electronic load you can use simple resistors (perhaps special glass-enclosed light-emitting resistors) and then adjust the voltage to get the desired current.
Or try different loads: .1 ohm, .2, .5, then 1 ohm, etc. and see at which point the problem begins.
Or wire up a big transistor that you feed a base current from another source.

Is the voltage regulation OK?
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #2 on: April 30, 2016, 05:42:28 am »
Are you measuring the current with a multimeter, or trusting the power supplies' inbuilt current meters? If you are using only the inbuilt meters, it could be either the power supply currents or the meters that have the problem.
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #3 on: April 30, 2016, 09:13:00 am »
Are there any ideas what would cause a maximum short circuit current of 30-50mA on two bench supplies?
Because this is junk with no CC mode. It just's cuts off the load at overload. I guess there is no proper heatsink too, just noisy fan cooled aluminium plate with fake 2N3055 which have very tiny silicon dies in them. This junk has nothing to do with mastech.
« Last Edit: April 30, 2016, 09:15:22 am by wraper »
 

Offline matt6ft9Topic starter

  • Contributor
  • Posts: 42
  • Country: us
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #4 on: April 30, 2016, 01:43:51 pm »
Voltage regulation in both power supplies works.  I have verified the voltage and current readings from the front panel meters with a DMM, the meters are working correctly.  The power supplies have only two power transistors on a backplate heatsink, no fan.

Jeroen79, thanks for the refresher, analog can be troublesome for me.  I'll have to look at the supplies more on monday.  I'll see if I can get some type of load together and maybe graph the output on both supplies.  I'll also poke around the inside with an o-scope again. 

Oddly enough, I have a Rigol DP832, but I like the MPJA bench power supplies for most applications.  The rigol has a little too many usage quirks.  (loud fan(yes, I know the fan could be changed), display shutting off after a short time, etc.)
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #5 on: April 30, 2016, 01:56:32 pm »
Are there any ideas what would cause a maximum short circuit current of 30-50mA on two bench supplies?
Because this is junk with no CC mode. It just's cuts off the load at overload. I guess there is no proper heatsink too, just noisy fan cooled aluminium plate with fake 2N3055 which have very tiny silicon dies in them. This junk has nothing to do with mastech.
If your supply has foldback current limiting, it could easily go down to 30mA with a short. If you set the current to maximum, and add a load below the maximum current, can you get the full current at the set voltage? So if you have a 10 ohm power resistor, set the voltage to 29 volts with the current limit at full, and see if you can get 29V across the 10 ohms for a few seconds.
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #6 on: April 30, 2016, 02:12:18 pm »
About meters. Check display accuracy while below 1V. In similar PSU I've seen ICL7107 clone used in some weird way with no negative supply and input voltage of the ADC referenced to GND at the same time. This caused significant nonlinearity at the bottom of the range because the input voltage was below the common mode range of the IC.
 

Offline matt6ft9Topic starter

  • Contributor
  • Posts: 42
  • Country: us
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #7 on: May 02, 2016, 07:27:18 pm »
Instead of digging further into the circuity, I decided a little sanity check was needed.  Output current and voltage data were taken trying to better characterize the issue being seen.  I borrowed a bunch of 50W, 1% resistors and took readings.  The power supplies were set at 12V, 5V, and 3.3V.  (see the spreadsheet below)  The short circuit current is still folding back to the 30mA to 50mA range.

I looked at another physically similar power supply, a 15 year old Mastech HY3003  (single output, 30V, 3A).   When the output is shorted with the fluke meter, the current draw goes to 2.07A.  (tested at 12V, 5V, and 3.3V)

A BK Precession 1735 power supply (single output, 30V, 3A) was also quickly evaluated.  When the output is shorted with the fluke meter, the current draw goes to 3.36A.  (also tested at 12V, 5V, and 3.3V)

Amspire, I think you’re right there might be some type of foldback current limiting on a direct short of the MPJA supplies.  It might be to keep the power supply from overheating?  I did set one of the power supplies to 29V, with a 10 ohm resistor, the current draw was 2.86A. 

So, maybe the MPJA power supplies are functioning as intended?
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #8 on: May 03, 2016, 01:14:10 am »
The supply is not meant to have foldback limiting, so something is wrong.

The supply is a rebranded Mastec HY5003 and someone else had a similar problem before, but no solution was mentioned:
https://www.eevblog.com/forum/repair/hy5003-psu-constant-current-mode-broken/

The current limit LED is on with the 50mA shorted output, and so something weird is happening around the LM741 current limit opamp (the top one in the circuits). Short the output and see what the voltages on pins 2 and 3 of the LM741 are compared to the positive output. Pin 3 should vary from 0V to maybe +0.5V depending on where the current limit is set. With a 50mA output, you would think pin 2 would only be something like +10mV, but to be in current limit, it would have to be at the same as pin 3. It should be the same at the voltage across the output 2N3055s' emitter resistors.

It is possible it is oscillating.
« Last Edit: May 03, 2016, 01:17:26 am by amspire »
 

Offline EPTech

  • Regular Contributor
  • *
  • Posts: 168
  • Country: be
    • EP Technical Services
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #9 on: May 03, 2016, 01:09:47 pm »
Hi there,

If you power supply works in the same way as the schematic you first posted, here are my ideas. Since voltage regulation is working perfectly, it means the bottom bottom opamp, the driver stage and the power stage are probably doing their job. I think the problem is situated near the current feedback, at the top opamp. The easiest way to test this is to desolder LED1. Do not short circuit the power supply because constant current will no longer work. Load the supply with ex. 0.5A using resistors or a load. After that, put the negative terminal of your DMM on the positive terminal of the supply which is the circuit ground reference. Measure the voltage on pin 2 of the top opamp. The voltage depends greatly on the shunt resistors but it will be a few 10th of a volt. Now measure on pin 3 of the opamp. You should be able to set the same amount of voltage using the potmeters on the front and the output of the opamp should toggle from the negative rail to the positive rail as you pass the value that is on pin 2 because we made it open loop by removing the LED. As an additional test you can hook the anode of the LED with a 2k2 resistor to the +12V rail. The LED should go ON and OFF as you pass the threshold on pin 2.
If you are unable to set a proper voltage on pin3, maybe VR2 is broken or the setting has drifted. If you are able to set the voltage but the opamp is not responding, maybe VR1 is broken or the setting has drifted. Else the opamp is dead. Note that if you replace the opamp, you need to set the offset again using VR1.

Good luck. I hope this helps.
Kind greetings,

Pascal.
 

Offline matt6ft9Topic starter

  • Contributor
  • Posts: 42
  • Country: us
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #10 on: May 04, 2016, 04:24:06 pm »
‘Finally found the correct schematic for the HY5003 power supply, see page 3 of the PDF here:
http://proto-pic.com/Datasheets/85-1858/datasheet.pdf
All of the component designators now match up to what I see on the PCB.  (the above PNG file has three components in faded yellow that are not stuffed on my PCB)

The 6.8K pot W6 and the 1.0K pot W8, was acting flakey.  (using the component references the above schematic)  The W6 was replaced with a 10K pot (closest I have, a 27K fixed resistor was added in parallel to the outer pins of the pot) and the W8 pot was shorted out. 

The -12V rail measured -15.8V.  So, V7A was replaced with a 12V Zener (1N4742).  The 12V, -6V, and -12V rails all look to be stable using an o-scope.  No change to the operation of the power supply.

I’m not sure what V17A (1N4148) and V18A(transistor) are doing.  When Vout is 12V, the anode of V17A is about at -4V.

Pascal, I tried what you said.  A lead of the V9 (the constant current LED) was desoldered.  If Vout=12V and there is no load, then pin 2 of N2(the constant current op-amp) is at 0V.  I set W6 (constant current pot) so that pin 3 of N2 is at 300mV.  Pin 6, the output of the op-amp N2, is at 11.6V.  When I attach a 10 ohm load, pin 2 of N2 goes to 510mV, pin 3 stays at the 300mV, and pin 6 goes to -5V.  So, it seems like N2 is functioning correctly.  I can leave the 10 ohm load attached, adjust the current pot up and down, and see the pin 6 output of N2 go between -5V and the 11.6V.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #11 on: May 05, 2016, 02:54:12 am »
What we want to know is what is happening on N2 when you have the current limit set to 3A, you have shorted the output and you are only getting 50mA out. The current limit LED is on so N2 must be limiting, but this should be impossible with 50mA out. At 50mA, the voltage on Pin2 should not be high enough to allow N2 to start regulating.

Probably has to come down to one of 5 things - (1) Something is affecting the Pin 3 voltage (2) something is affecting Pin 2 voltage (3) N2 is oscillating (4) N2 has a power supply problem (5) something is wrong in the zero offset circuit of N2. If you can find out which it is, it will be easier tracking down the fault. The fact that the same fault is in two supplies, it implies that the problem is in something about the construction design, or a chronic fault in a certain component.

One area of concern it that the zero offset pot normally is not directly connected to the -6V rail. This means that if this pot is at one or extreme or other the 741 cannot function properly at all. You would normally have something like the 10K pot with a 4k7 resistor each side.

Richard
 

Offline Kevin.D

  • Frequent Contributor
  • **
  • Posts: 290
  • Country: england
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #12 on: May 05, 2016, 01:48:56 pm »
It's foldback current limiting .When the anode of V17 is at ~1.3V relative to aux gnd.
V18 is turned on and pulls down the current limit setting  .
The ratio of R17/R19 determines at what value of output voltage fold-back current limiting
will become active. Also this has some hysterisis so that once activated the load has to be removed
to reset full current limit. Using the two values for R17/R19 given in your schematic it's currently set
 to become active about 6V out. It's odd since you wouldnt normally see any foldback current limiting on a bench supply because your supossed to design the supply to be able to source the full current range down to it's minumum output voltage setting(usually 0V) , so any foldback limiting simply negates your minimum working voltage range at full current. If you dont want it Lift R17 to disable it or up R17 to lower active threshold (R17 -10k sets threshold to ~.8V) .


Regards
 

Offline matt6ft9Topic starter

  • Contributor
  • Posts: 42
  • Country: us
Re: MPJA Benchtop power supply, only 30mA-50mA short circuit current
« Reply #13 on: May 05, 2016, 08:25:37 pm »
Interesting, that makes sense.  As a check, I lifted a leg of R17 and now when I short out the output, the current goes to about 3.1Amps. But, I plan on putting it back down for normal operation.  (I could wire the disable to a switch on the back of the power supply, if needed)
Amspire, kevin.d, eptech and others:  Thanks for all the great help, it’s been a learning experience. 

Summary for those pursuing this thread: 
The MPJA 14602PS (HY5003) has a built in current limit of about 30mA to 50mA when the power supply output is shorted.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf