Bench Power Supply

[Electro Fan]

fwiw,

Here is a current measurement showing a Korad KA3005P reading as it powers a simple DUT vs readings from an Aneng 8008, a Fluke 179, and a Keysight 34465A.  (They were all strung together in a circuit.)

Edit:  a couple observations.  1) I think the Korad is good to within about a milliamp, 2) might have to tune up the 34465A   (To be square, I think by the time I got to the 34465A the current had been drifting/settling slightly downward which pulled the average down a tad - but now we are talking about my ever changing air conditioned temperature and we are getting into volt nut, temp nut, etc nut territory.)  Bottom line:  If you want a power supply and your budget is ~$100, just buy a Korad KA3005D and keep moving toward whatever your goal might be.

[radiolistener]

It might be worth checking out how accurate the meters on the Korad supply are. Recently I bought a bunch of Gophert PSUs (these are switching ones) from Aliexpress but their meters are shockingly accurate. Especially given the price.

Here is my test for KORAD KA3005D from another thread:

Unloaded output.
Format is [setting voltage] = [measure output]:
0.10 V = 0.10159 V
0.50 V = 0.49860 V
0.70 V = 0.69990 V
1.00 V = 0.99695 V
1.01 V = 1.00594 V
1.20 V = 1.19851 V
2.00 V = 1.99745 V
3.00 V = 2.99477 V
3.20 V = 3.19320 V
4.00 V = 3.99448 V
5.00 V = 4.99893 V
6.00 V = 5.99988 V
7.00 V = 6.9970 V
8.00 V = 7.9970 V
9.00 V = 9.0025 V
10.00 V = 9.9969 V
11.00 V = 11.0012 V
12.00 V = 11.9992 V
13.00 V = 12.9997 V
14.00 V = 14.0037 V
15.00 V = 15.0026 V
16.00 V = 16.0053 V
17.00 V = 17.0020 V
18.00 V = 18.0143 V
19.00 V = 19.0084 V
20.00 V = 20.0103 V
21.00 V = 21.0092 V
22.00 V = 22.0081 V
23.00 V = 23.0159 V
24.00 V = 24.0120 V
25.00 V = 25.0161 V
26.00 V = 26.0132 V
27.00 V = 27.0226 V
28.00 V = 28.0181 V
29.00 V = 29.0227 V
30.00 V = 30.0234 V
31.00 V = 31.0211 V

Loaded output test.
Format is [voltage setting] => [measure unloaded voltage] / [measure loaded voltage] @ [measure load current]
3.20 V => 3.19293 V / 3.17939 V @ 4.138 A
5.00 V => 4.99890 V / 4.98319 @ 4.803 A
9.00 V => 9.0027 V / 8.9877 V @ 4.630 A
12.00 V => 11.9991 V / 11.9808 V @ 4.703 A
19.00 V => 19.0084 V / 18.9909 V @ 4.798 A
24.00 V => 24.0118 V / 23.9928 V @ 4.751 A
30.00 V => 30.0231 V / 30.0025 V @ 4.809 A

Measurement is done with Brymen BM867S.

KORAD allows to perform calibration. And I done precise calibration when I bought it.

[radiolistener]

KORAD KA3005D current measurement and stabilization test.

Format: [current setting] => [multimeter measure current] / [KORAD measure current]

0.001 A => 1.540 mA / 0.000 A
0.002 A => 2.744 mA / 0.000 A
0.005 A => 5.992 mA / 0.003 A
0.010 A => 10.860 mA / 0.009 A
0.015 A => 15.858 mA / 0.014 A
0.020 A => 20.631 mA / 0.019 A
0.030 A => 30.434 mA / 0.028 A
0.040 A => 40.916 mA / 0.038 A
0.050 A => 50.704 mA / 0.049 A
0.100 A => 100.75 mA / 0.099 A
0.150 A => 150.30 mA / 0.148 A
0.200 A => 200.86 mA / 0.199 A
0.250 A => 250.98 mA / 0.249 A
0.300 A => 300.59 mA / 0.298 A
0.350 A => 350.47 mA / 0.349 A
0.400 A => 400.59 mA / 0.399 A
0.450 A => 450.05 mA / 0.448 A
0.500 A => 500.62 mA / 0.499 A
0.550 A => 0.5507 A / 0.549 A
0.600 A => 0.6003 A / 0.599 A
0.700 A => 0.7009 A / 0.699 A
0.800 A => 0.8019 A / 0.800 A
0.900 A => 0.9014 A / 0.900 A
1.000 A => 1.0024 A / 1.001 A
1.500 A => 1.5026 A / 1.500 A
2.000 A => 2.0012 A / 1.999 A
2.500 A => 2.5024 A / 2.500 A
3.000 A => 3.0031 A / 3.000 A
3.500 A => 3.4987 A / 3.498 A
4.000 A => 3.9991 A / 3.998 A
4.500 A => 4.5016 A / 4.500 A
5.000 A => 5.0005 A / 4.999 A
5.100 A => 5.1007 A / 5.100 A

Bought this unit almost year ago for about 80 USD include shipping cost

In comparison with Dave teardown video, my unit is almost the same. But there is a difference for power regulator module. Heatsink is now more massive with better transistors mounting.

[Yansi]

That is some of the very newer designs. Mine is much older, also has three snap-in caps in parallel. The photo is from January 2014, when I've bought my second Korad PSU and was modifying the hilarious heatsink for a proper one, so the fan can shut the fuck up.

Also I've never done any calibration to any of my Korad PSUs, would be interesting to test, have a calibrated 34401 available now so...

[tooki]

That is some of the very newer designs. Mine is much older, also has three snap-in caps in parallel. The photo is from January 2014, when I've bought my second Korad PSU and was modifying the hilarious heatsink for a proper one, so the fan can shut the fuck up.

Also I've never done any calibration to any of my Korad PSUs, would be interesting to test, have a calibrated 34401 available now so...

I really need to do that. I put a better, quieter fan in, but the janky heat sink makes it moot.

[Croby80]

Hello, I'm looking for a good (but not expensive) power supply unit for my personal laboratory and I'm observing by long time the following model on aliexpress store:

https://it.aliexpress.com/item/1005001615696493.html?spm=a2g0o.cart.0.0.5a3d3c00R6iaCI&mp=1

it is a single channel 300W programmable power supply with a declared accurancy of 0,01mA for current and 0,1mV for voltage with capability to record and graphically visualize voltage current and power . It is possible to define 100groups of timed programmed steps and also to manage battery charge slopes. RS232, analog and USB interface are available (LAN optional) to be conneted to a PC to remotly control and to visualize V,I,P,R parameters. Sense compensation is another feature with a 4,3" rgb graphical display.  Price = 511euros.

Brand is AB-F but I suppose could be a rebrand for eTommens manufacturer: http://www.etommens.com/dgweb_content-847183.html

Anyone has experience with these devices?  What do you think about that?

(I attached datasheet of previous 5 digit version =0,1mA, 1mV)

Thank you

[tautech]

Hello, I'm looking for a good (but not expensive) power supply unit for my personal laboratory and I'm observing by long time the following model on aliexpress store:

https://it.aliexpress.com/item/1005001615696493.html?spm=a2g0o.cart.0.0.5a3d3c00R6iaCI&mp=1

it is a single channel 300W programmable power supply with a declared accurancy of 0,01mA for current and 0,1mV for voltage with capability to record and graphically visualize voltage current and power . It is possible to define 100groups of timed programmed steps and also to manage battery charge slopes. RS232, analog and USB interface are available (LAN optional) to be conneted to a PC to remotly control and to visualize V,I,P,R parameters. Sense compensation is another feature with a 4,3" rgb graphical display.  Price = 511euros.

Brand is AB-F but I suppose could be a rebrand for eTommens manufacturer: http://www.etommens.com/dgweb_content-847183.html

Anyone has experience with these devices?  What do you think about that?

(I attached datasheet of previous 5 digit version =0,1mA, 1mV)

Thank you

Just what are your real requirements ?
Precision or raw power ?

[Yansi]

That is some of the very newer designs. Mine is much older, also has three snap-in caps in parallel. The photo is from January 2014, when I've bought my second Korad PSU and was modifying the hilarious heatsink for a proper one, so the fan can shut the fuck up.

Also I've never done any calibration to any of my Korad PSUs, would be interesting to test, have a calibrated 34401 available now so...

I really need to do that. I put a better, quieter fan in, but the janky heat sink makes it moot.

The original fan is quiet enough, if you put a resistor in series. I gues it was something like 33 or 39 ohms, what I've added there.

//EDIT: 68ohms, according to my photos.

[Yansi]

Hello, I'm looking for a good (but not expensive) power supply unit for my personal laboratory and I'm observing by long time the following model on aliexpress store:

https://it.aliexpress.com/item/1005001615696493.html?spm=a2g0o.cart.0.0.5a3d3c00R6iaCI&mp=1

it is a single channel 300W programmable power supply with a declared accurancy of 0,01mA for current and 0,1mV for voltage with capability to record and graphically visualize voltage current and power . It is possible to define 100groups of timed programmed steps and also to manage battery charge slopes. RS232, analog and USB interface are available (LAN optional) to be conneted to a PC to remotly control and to visualize V,I,P,R parameters. Sense compensation is another feature with a 4,3" rgb graphical display.  Price = 511euros.

Brand is AB-F but I suppose could be a rebrand for eTommens manufacturer: http://www.etommens.com/dgweb_content-847183.html

Anyone has experience with these devices?  What do you think about that?

(I attached datasheet of previous 5 digit version =0,1mA, 1mV)

Thank you

For abotut the same price you can get a second hand 300W HP/Agilent PSU. This I'd never pay so much for, especially if it's something noname, and likely a switchmode noise transmitter.

[Elasia]

I've been looking at this 1800W A-BF... only reason i havent bought it is because I fear burning the house down with it 

https://www.aliexpress.com/item/4001126735505.html?spm=a2g0o.productlist.0.0.5e891a6f9Q3V1a&algo_pvid=99b7ccc9-cb97-4a59-8de1-e2ca6a5ae7a4&algo_expid=99b7ccc9-cb97-4a59-8de1-e2ca6a5ae7a4-4&btsid=0bb0623216068558525863252ed93f&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

Any news with what Siglent is doing with that 5000 series yet tau? Been eyeballing one of those but seems like a year wait

[nctnico]

Hello, I'm looking for a good (but not expensive) power supply unit for my personal laboratory and I'm observing by long time the following model on aliexpress store:

https://it.aliexpress.com/item/1005001615696493.html?spm=a2g0o.cart.0.0.5a3d3c00R6iaCI&mp=1

it is a single channel 300W programmable power supply with a declared accurancy of 0,01mA for current and 0,1mV for voltage with capability to record and graphically visualize voltage current and power . It is possible to define 100groups of timed programmed steps and also to manage battery charge slopes. RS232, analog and USB interface are available (LAN optional) to be conneted to a PC to remotly control and to visualize V,I,P,R parameters. Sense compensation is another feature with a 4,3" rgb graphical display.  Price = 511euros.

Brand is AB-F but I suppose could be a rebrand for eTommens manufacturer: http://www.etommens.com/dgweb_content-847183.html

Anyone has experience with these devices?  What do you think about that?

(I attached datasheet of previous 5 digit version =0,1mA, 1mV)

Thank you

For abotut the same price you can get a second hand 300W HP/Agilent PSU. This I'd never pay so much for, especially if it's something noname, and likely a switchmode noise transmitter.

But not a high precission one. With a weight of 7kg it seems to me there is a regular transformer inside these power supplies with a linear power stage. I don't think you can get the low noise and accuracy from a switcher so it has to be linear. Or at least have a linear power stage behind a pre-regulator.

I've been looking at this 1800W A-BF... only reason i havent bought it is because I fear burning the house down with it 

You can also look for a Gophert PSU. These seem to have a good track record.

[tautech]

Any news with what Siglent is doing with that 5000 series yet tau? Been eyeballing one of those but seems like a year wait

Nuthing to report as I've not asked but will today and post anything of interest in the SPS5000X thread.

[Datman]

I have an ATTEN APS3005Si. It a linear PS (no switching noise at the output). I have replaced the 4 pots with 2 multi-turn, added a thermostat and a resistor and a trimmer to adjust Ammeter offset. I have replaced the 470uF output capacitor with a 10uF one, avoiding current peaks in CC mode.
Look at this thread:
https://www.eevblog.com/forum/reviews/atten-aps3005si-pots-replacement-and-meters-adjustment/

[Croby80]

I need an accurate PSU with almost 200W of power. I attched the internal power supply composition, I suppose it could be a switching + linear regulation method as declared by custumer for the previous less accurate model  SS-L3010SPL (about 360 euros). 

[tautech]

I need an accurate PSU with almost 200W of power.

Is 150W enough ?
https://www.siglenteu.com/power-supplies/spd1000x-series-programmable-dc-power-supply/

[Croby80]

Better 200W, one single output is not a problem

[Croby80]

I'm comparing DP832 vs SPD series programmable supply, I see Rigol is more accurate respect to Siglent device but more expensive. I also seen that Rigol is able to generate prefixed output waveforms such as sine, ramp or steps defining their characteristic parameters: are these waveform completely analog or discrete? I mean, the 2048 (V,I) samples sets of Rigol define the maximum resolution of the output waveform?     

[Yansi]

I need an accurate PSU with almost 200W of power. I attched the internal power supply composition, I suppose it could be a switching + linear regulation method as declared by custumer for the previous less accurate model  SS-L3010SPL (about 360 euros).

Clearly tthere is no switchmode pre-regulator, instead, in the bottom part of the photo, there are two relays marked as "group switching"  (selecting the secondary voltage from 4 possible configurations).

[tautech]

I'm comparing DP832 vs SPD series programmable supply, I see Rigol is more accurate respect to Siglent device but more expensive.

Which model ?
These ?
https://www.siglenteu.com/power-supplies/spd3303x-spd3303x-e-series-programmable-dc-power-supply/

[uski]

(and I'm not adverse to adding a voltage divider if finer adjustment is required)

What? If you think about a resistive voltage divider at the output, you most likely cannot do this due to the equivalent source impedance. Or were you referring to adding a separate potentiometer to the existing potentiometer for finer voltage adjustment ?

The PSU is going to be an xmas present (hence the limited budget).

This is exactly how I got my first sh*tty power supply that I regretted purchasing. My parents bought it for my as a christmas present years ago.
I'm not saying to buy something expensive, but please do yourself a favor and buy something good. See my suggestions for the Siglent PSUs.

Remember, these instruments will last you for years. It makes sense to invest a bit more in the beginning and have a good instrument from the beginning, instead of suffering from a poor instrument and then having to pay for a better one later anyway. I'm trying to save you from disappointment that I experienced in very similar conditions years ago.

Specifically the one I had had horrible regulation characteristics where it could not setup the current limit down to 0, there was a dead band between 0 and 200mA or so, and also, at these voltages, the voltage output changed suddenly around the current limit setting. I hated that power supply !!!

An alternative if you are cash strapped is to buy a single but good quality power supply, and then make yourself an adjustable power supply with an LM317 regulator. It's a fun project, it will be cheap, and there will be many situations where you will not need much power, for instance when powering digital circuits. Or go fancy with a UA723 and add current limiting and so on.

PS: regarding the Siglent PSUs, they can be hacked to get more precision, if you are really cash strapped. SPD3303X-E with 10mV/10mA precision can be hacked to a SPD3303X model with 1mA/1mV. Not commenting on the ethics/legal/warranty aspects of that modification, it's a tradeoff you have to consider yourself. But in my view, a hacked SPD3303X-E is infinitely better than a crappy PSU.

[Croby80]

Yes, the SPD3303X model

[tautech]

I'm comparing DP832 vs SPD series programmable supply, I see Rigol is more accurate respect to Siglent device but more expensive.

Which model ?
These ?
https://www.siglenteu.com/power-supplies/spd3303x-spd3303x-e-series-programmable-dc-power-supply/

Yes, the SPD3303X model

Then you need study what uski wrote 

PS: regarding the Siglent PSUs, they can be hacked to get more precision, if you are really cash strapped. SPD3303X-E with 10mV/10mA precision can be hacked to a SPD3303X model with 1mA/1mV. Not commenting on the ethics/legal/warranty aspects of that modification, it's a tradeoff you have to consider yourself. But in my view, a hacked SPD3303X-E is infinitely better than a crappy PSU.