Calibrating powersupplies

[Rob Sims]

Hi,

My Tekpower TP3005T powersupply is off in the higher voltage ranges. It is a 0 - 30V powersupply. From 0.000V to 10.00V it is 0.001V higher than the reading on my calibrated DMM. From 11.00V to 16.00V it is 00.02 higher than the reading on my DMM. From 16.00V  to 30.00V it is 00.06V higher than the reading on my calibrated DMM.  I want to set the voltage reading of the powersupply according to the DMM. Can the voltage reading of a powersupply be adjusted or not?. If so can someone who maybe did this before, tell me how to do it , what to look for to do I etc..??

Many thanks in advance.

[Vgkid]

You are a fraction of a percent out. Not enough to be worried about, but if you want to change it, look into the pots on the display board.
it is specced at 1% so not likely. Is there anything under the front of the display?

[w2aew]

You are a fraction of a percent out. Not enough to be worried about, but if you want to change it, look into the pots on the display board.
it is specced at 1% so not likely. Is there anything under the front of the display?

Agreed - too small to really be concerned about.  You're likely to drop this much voltage across the leads between the supply and the circuit under test, unless you're remote sensing.  In fact, if you are not remote sensing your supply, then you should always be in the practice of measuring the supply voltage at your circuit anyway so that you can account for any IR drop on the wires to the supply.

[Rob Sims]

Many thanks guys,

The current readings readings of the powersupply are spot on with my calibrated multimeter. But how do I know which pot to adjust for voltage???

 

[jeroen79]

Find a service manual.

[BradC]

Many thanks guys,

The current readings readings of the powersupply are spot on with my calibrated multimeter. But how do I know which pot to adjust for voltage???

Easy. Use a pattern search. Just gently nudge every pot you find until you find the one you want to change. This has been the method used by celebrated knob twiddlers since the invention of the knob and never fails to identify the desired unit.

[Rob Sims]

Thanks Brad C. I will open up the box tomorrow and look for the pots. Hope there is only one pot for voltage and  one for current.

[eKretz]

If you do this, be sure to mark the exact location of each pot before it's turned so you have a reference to return to - otherwise you'll be recalibrating everything.

Signed - a fellow knob twiddler.

[Rob Sims]

Thanks for the advice eKretz. It also came across my mind to mark the locations, because recalibrating everything is not what I want. Tomorrow I will open up the box and look for the pots.

[BradC]

Sorry, I was being highly facetious. I can guarantee if you follow my method you'll do more damage than good. Poking around the pots with a meter is probably a good start but don't go shoving pots Ess you are prepared to spend a great deal of time undoing the damage.

[Kilrah]

Given that it's pretty much perfect as it is, it's almost guaranteed that without access to the manufacturer's adjustment procedures (service manual) and equipment you'll end up worse.

[Mechatrommer]

that random poke advice should be ban of the year... if you poke something that matched the other side of a circuit like diff amp input matching, no multimeter reading or marker position lock will do the job to restore your sorry ash...

[2N3055]

I would not touch a thing.. 0.06 V at 16V is max error of cca 0,375%, otherwise between 0.1 to 0.2%.
That is perfect for this class of equipment, and much better than specified (if I read correctly manufacturer promise something arround 2%!!!)

60 mV error on 30V power supply is a non issue...

[Rob Sims]

Thanks 2N3055. Woke up this morning with a bad feeling that my power supply won't be the same if I start turning pots. I looked at the specs again and the manufacturer states that LCD Accuracy is within +/- 2.5%. I checked the readings on the power supply again against my fresh calibrated DMM.

From 01.00 to 10.00 V     error 0.001
From 11.00 to 16.00 V     error 00.02
From 17.00 to 21.00 V     error 00.04
From 21.00 to 32.00 V     error 00.06

I’m a mechanical engineer, so I consider myself a newbie in electronics. My main goal with electronics is to design control circuits for mechanical devices.   I agree that the power supply accuracy is well below manufacturers accuracy, but my concerns are :

A If this error range  isn’t a bit too high for circuit design with precision electronic components
B This powersupply is  chinese. Bought it via Amazon. If the readings further drift later I don’t want to
   ship it again, and I don’t want to send it to a electronic repair shop. Since I’ve started with electronics  I
   want to calibrate or repair  it myself.

Hope someone who calibrated a power supply before can give me some guidelines.
       

[Kilrah]

A If this error range  isn’t a bit too high for circuit design with precision electronic components

No. When you want an ultrastable supply for a circuit then you design a specific power supply for that circuit to your own spec, you don't rely on whatever is upstream. If you need ultra clean 12V you'll design your system to take the already pretty clean 15V input from that power supply, and regulate that to your own ultra clean 12V onboard as per your spec and with it being fully under your control.

Yes you can get ultra-precise supplies you could use straight, but then they're not in the same class and will cost a couple of orders of magnitude more.

[Rob Sims]

No, it is not about a stable supply. The Tekpower TP3005T Powersupply I have is very stable. It's about the readings from the display of the PSU. I want the readings from the display to be in sync with the reading on my fresh calibrated Digital Multimeter. I want someone to tell me how to adjust the readings of the PSU so it matches the readings of my DMM. The powersupply is a digital supply, not a continuous variable powersupply. I need somenone who has experience with these Chinese supplies to give me some guidelines.   

[GBowes]

Surely as a mechanical engineer, you understand tolerances.

You say that the spec on the LCD display is ±2.5%. You report a maximum display error of .060V in 21.00V (An error of + 0.29%)
With which part of "It is well with the mfr tolerance specification" are you having trouble?
If there is an adjustment for the display, which I doubt, it it unlikely that it is sensitive enough for you to improve on this.

If it is really so critical to you, and the output is accurate enough for you, then why not spend a few thousand dollars and purchase a 0.2% Meter and you will reduce your display error to 0.042V

[Zero999]

B This powersupply is  chinese. Bought it via Amazon. If the readings further drift later I don’t want to
   ship it again

1) The readings could drift in either direction. It could just as easily get better as get worse.

2) If it drifts considerably out of tolerance, i.e. more than 2% or so, then you can adjust it later.

3) As many people have stated above: there's a higher chance you could make it worse than better.

Did you leave it running for an hour or so to warm up, before checking it?

[Rob Sims]

Good point. Yeah I think i might be overreacting. But it would be nice to hear from someone who has calibrated their own power supply. All the small panel voltage and current meters sold at Amazon have small calibration trim pots to set voltage and current. I doubt it that a $1000 PSU doesn't have at least one trim pot. 

[Rob Sims]

Hi Hero999

The supply is very stable. If I leave it running the whole day the readings stay the same.

[2N3055]

Rob,

many people here, me included, calibrate all kinds of equipment on regular basis... When it needs calibration..
Your PSU doesn't.

I would like to go back to begining: Please specify exact make and  model of multimeter you used to check PSU and what was last date of calibration of said multimeter...

If that multimeter is not at least 4.5 digit and better than 0.05 % accuracy, and calibrated to be sure it's within specs, you might be measuring meter's error.. 
Even good brands 4000 and 6000 display meters are usually 0.5 %...

Or, you being mechanical engineer,let me use your lingo: you must check vernier calipers with micrometer... Other vernier caliper won't do for check of 0.05 mm accuracy.. With other calipers you can only check for gross error...

Or simply, if error was 0.4 volts, PSU would be surely out of specs and would need calibration.. With what you have now, and if you don't have sufficiently accurate equipment and experience, you'll probably make it worse..

Also, if PSU is brand new, I would leave it alone for few months for initial drift to settle and then worry about making it more precise..

All the best...

Sinisa

[bitseeker]

Hi Rob,

Welcome to the forums. Yes, you are overreacting a bit, though it's not unusual since these things are a new experience. Following are some things to be aware of when it comes to adjusting power supplies and hopefully the information will improve your comfort level:

• The difference in output that you're seeing is a limitation of the power supply measurement and display system. It's not as good as your DMM. To get better resolution and/or accuracy requires a better meter in the power supply, which would increase its cost.
• Even with a better meter in the power supply, as Alan pointed out, you will still need to measure the voltage at the point in your circuit where you're applying power because your supply doesn't have remote sensing. The actual voltage at your circuit will be less than is on the display due to cable resistance. The amount of loss will depend on the amount of resistance in the cables (which increases with length) and how much current your circuit is drawing. Ultimately, the display on the power supply is for convenience.
• If you can find a schematic, the adjustment potentiometers will be identified. However, there may be specific steps required to accomplish a proper adjustment, which requires a service/calibration manual.
• Unfortunately, it's often difficult to get service information, let alone a schematic, of mass-market power supplies from China. Sometimes you can get lucky. Many TekPower supplies are similar to Mastech, Volteq, and a bunch of other brands. I'm not sure if they're all identical, but you might able to find a schematic when searching under a different name.
• If there are adjustment potentiometers, unless there are separate ones for different ranges, you'll usually lose accuracy on some ranges when gaining accuracy on others. As a result, you end up having to find a position that's good enough across them all.
• I have several high-end power supplies. They don't have potentiometers. They're all digitally/programmatically adjusted (a.k.a., closed case calibration). So, not all power supplies have them even when they can be adjusted.
• Feel free to explore how an adjustment can be done on your power supply, but as everyone has confirmed, don't worry about doing anything right now. It's operating just fine.

[bitseeker]

I found some photos of the inside of the 3005T in this thread: https://www.eevblog.com/forum/testgear/tekpower-tp3005t-variable-linear-dc-power-supply/msg719478/

The sixth photo is of the front panel board. At least on the side that was photographed, I didn't see any adjustment pots. They may be on the other side of the PCB. However, since this is a digital supply, it might have calibration constants stored in non-volatile memory.

[Rob Sims]

Thanks Bitseeker. I found the seller of these units. I will send them a email. If I have time I will open the unit to look for pots myself. However as you said, it's a digital unit so maybe there a no pots to find.

[Rob Sims]

Thanks Bitseeker. I didn't see your first post. Yes you're right.  I'm convinced now.