Yihua 305D, 3010D.. Myth?

[tarmogr]

Hi, I'm a Yihua vendor in Europe and I just saw this topic, claiming unacceptable voltage spikes and unstable stabilization.
https://www.eevblog.com/forum/reviews/beware-yihua-yh-305d-bench-psu
I have some power supplies in stock myself and decided to test if this is true.

Maybe they have upgraded their design or this Youtube user got one from a bad bach. Aniway I got completely different result:

Switch-on response (Yihua PS-3010D)


Automotive bulb hooked up, also tested it with a resistor and diodes in series, like in the original video, I got excactly the same result.


Switch-on response (Yihua PS-305D)


It looks pretty smooth to me. Peak-to-peak voltage under ~0.5A load was about 60mV, measured with my Rigol DS1052E (unmodded). I could't get the resolution low enough to measure it accurately.

[amyk]

How many did you measure? A sample size of 1 isn't particularly conclusive.

As you mentioned it's been more than a year since those issues were reported so maybe revisions to the design have solved them already.

[alm]

To answer your title: no, you can never discount the outcome of an experiment as a myth. You might dispute the interpretation, or the way the experiment was performed, but not the result.

It may have been an older batch. It may be a sign of poor QC (you need a huge size from several batches from different time periods to tell this).

The setup might also have been different from yours. He uses three heavy-duty diodes as load. A light bulb is going to represent a very different load (an increasing resistance starting with a very low value). Neither are anywhere near resistive, but a lab supply is supposed to perform well in a huge variety of loads: how many of your circuits consist of only resistors?

The charge left in the caps might also make a difference. I have one supply that will initially deliver the charge stored in the output cap (the previously set voltage) until this cap is discharged, from that point on it will deliver the set voltage (note that this is a specialty precision power supply, not a general purpose one, so I don't really care). So plugging in with empty caps, setting to 0V, switch off and switch on and set to 20 V, switch off and switch on may all have different outcomes. Most supplies will discharge the caps when powered off, so unless you're quickly cycling them this shouldn't be an issue.

[tarmogr]

How many did you measure? A sample size of 1 isn't particularly conclusive.

As you can see I tested 2 different powes supplies. One is manufacture in July, 2013 (3010D), the other one is about 2 years old (305D)

I tried to keep the variables as close to the original test as i could (0,5A, ~2,8V) and also put a 6.8 ohm resistor in series with 3x 1N4007 diodes (not heavy duty I know) to replicate the original test, the results were basically identical to the tests above.

I have a total of 13 supplies in stock from different baches. If you wish I could do further testing with diodes, try to discharge the caps and maybe increase the load, what do you think?

[Dago]

I have a same type of PSU (different brand but looks identical, has LCD displays instead of LED though) and it doesn't have any voltage spike on turn-on either.

[JimmyMz]

tarmogr,

Do you sell the Yihua PS1502DD? I have a wire that came loose from the mainboard. I don't have any idea where it connects, and there is no schematic available for this specific model. It's not one of the wires that comprise the wire harnesses; it's simply a single wire that jumps the mainboard to the display board. I would be very appreciative as the seller never responded (a year ago), and I want to give this power supply to a friend. If you have one, and don't mind spending 10 minutes, I would be thankful.

[metalphreak]

The Vantek supply I have is fine with turn on voltage, but others with identical supplies (under a different brand) have had turn on spikes. Pictures posted of the internals showed them to be basically identical. At the end of the day, it's probably a firmware issue on those because they are programmable. Do these basic supplies even have digital control?

[alm]

I doubt that it is firmware. The problem is most likely to occur on power up before the firmware gets a chance to intervene and all pins are high-impedance. This is the hardest part to get right.

[electronupdate]

The Yihua 3010d has an analog control loop built out of op-amps.  I would not be surprised that some units may be better than others due to component tolerances

The one I got (from about Feb of this year) definitely showed poor stability  esp. when turning on.... it was not particularly difficult to get the supply to generate some rather large voltage output spikes if you cycle the power 'just right'.

Good to hear that the unit you tested from a July 2013 is more stable.

[Sylwerdragon]

I was thinking about this power supply...but after few movies about it... I give up...I bought some isotech...and that is the worst power supply ever.. So i will try to buy some good one even if it will cost a lot more than normal CH.

[Tronto]

Hi;

Do you happen to have a schematic for theYihua PS-305D?   Mine came with a blown resistor R144 and now won't attenuate the voltage at all, just sits at 49.3 volts 6.4 Amps CV.  R144 feeds the LM324N.

Cheers