Is My DC Power Supply Garbage?

[malekia]

Hi, I'm new to this forum. I've spent a lot of time searching and reading but haven't found the answer to my question.

I bought a $70 power supply off EBAY and though my expectations were low I feel that this may not be worth keeping. The PS looks like many others (same case) on EBAY and even a few on TEquipment (Elenco).

The reason I'm apprehensive about this unit is that it's incredibly light (a little over 4 lbs). Also, I've seen several teardowns of similarly priced "EBAY" power supplies and in every case they had much larger transformers than mine.

I want to keep the price below $200 and I'm considering returning this unit and buying one of these:
- Circuit Specialists PPS2320A
- B&K Precision 9110
- Velleman PS603U
- Unisource PS-1305

I don't have any specific requirements; I'm new to this. I just don't want to buy complete.

Any idea would be appreciated.

Ali

[stmdude]

You know what?   It's probably not terrible.

I have one that looks just like it, but different branding (seems this model is heavily OEMd), and it works okay. Better than a few other supplies that I've tried that have issues responding quickly to quick changes in current-draw.

For bring-up stuff, I use an old high-end HP PSU (finer control of the CC limit), but when I just need 3.3V, etc, this is my go-to PSU. Simple, and works as advertised.

Inside yours looks typically chinese. They've done cost-cutting, but the only obvious thing wrong I see after a quick glance would be the naked solder-tabs on the ON/OFF switch, which would be live mains. That should be covered with heat-shrink at least.

Apart from that, test it out before you get rid of it. Put some load on it, and look at the ripple (if you have a scope), and how precisely it's calibrated with regards to voltage and current-limiting with a multimeter.

[Fungus]

It could be OK. Check out the voltage ripple under load, make sure there's....if it looks OK then use it.

The reason it's light is because it's a switch mode supply, not a big honking transformer.

[Mark]

It has a "QC Passed" sticker... it MUST be good    

[mmagin]

It has a "QC Passed" sticker... it MUST be good    

It's the gold color that makes it legit! 

Seriously though, this probably doesn't have the cleanest output, but it probably works okay.  I would be inclined to make sure there's no DC path from line or neutral to the output side for safety reasons, but I'm cautious.

[newbrain]

The reason I'm apprehensive about this unit is that it's incredibly light (a little over 4 lbs). Also, I've seen several teardowns of similarly priced "EBAY" power supplies and in every case they had much larger transformers than mine.

I would not like to state the obvious, but have you considered that this looks like a switching PSU, but the ones with the beefier transformers were most likely linear PSUs?

[Kleinstein]

It's a switched mode supply, so expect more high frequency noise than with a heavy linear supply. If it is a pure switched mode supply, also the speed of regulation is often poor.

I don't really like the heavy heat sink on a the board - this might break things on shipping. Also isolation does not look really good, though not really bad spot visible. But a lot of cables hanging around and parts of the board are mains connected. E.g. the thin cables near the power switch are not good, though not dramatic. Its also a big question if the large heat sink is for the switcher and thus potentially life of if there is a linear stage behind it.

[malekia]

It has a "QC Passed" sticker... it MUST be good    

It's the gold color that makes it legit! 

They should put a few of those stickers on the outside 

[malekia]

Thanks for all of the replies.

I decided to return this and bought a very new-looking Agilent E3610A on EBAY.
Before I send the other one back I made an attempt to measure ripple. This is my first attempt so no laughing please!

I followed as best I could EEVblog #594 - How To Measure Power Supply Ripple & Noise

Scope: Rigol DS1054Z
Coupling: AC
B/W Limit: 20MHz
Probe Ratio: 1x
Rigol RP2200 probe specs - Bandwidth: 1x: DC~7MHz
Acquisition: Peak detect Mode
Trigger: Auto

I don't have an electric load device so I connected an RC LiPo charger and measured while charging. Also, I don't have a probe to BNC adapter so I connected the probe directly to +/-.

PS Display: 27.1V, 1.2A

Please see photo for results.

[mmagin]

You probably have weirdness from the switching regulator in the LiPo charger too.

I'm quite happy with my E3610A.  Seems generally as good as my 1970s HP supplies, but easier to read.  A caution: the metering was out of spec on mine -- the voltage is slightly off, and the current limit setting button doesn't make it show the correct current limit at all, it's way off of what it actually limits to.  I suspect adjusting those into spec won't be that hard, but I wanted to warn you not to trust it until you've checked it with a DMM.

[stmdude]

A caution: the metering was out of spec on mine -- the voltage is slightly off, and the current limit setting button doesn't make it show the correct current limit at all, it's way off of what it actually limits to.  I suspect adjusting those into spec won't be that hard, but I wanted to warn you not to trust it until you've checked it with a DMM.

Yup, you should definitely do this.

Also, the E3610A is a really nice PSU. I used one at work up until a few years ago (I moved to an office in a different country, PSU didn't get to come with). I had no complaints at all, and I'm kind of picky with my PSUs

[Kleinstein]

For measuring the noise from the PSU, one should have a simple resistive load, like a power resistor or a simple incandescent lamp. Other loads might add extra noise.

[malekia]

My Agilent E3610A arrived yesterday and I have been trying Kleinstein's incandescent light bulb idea but haven't been able to light a 25w bulb.

Remember that this is a 30w PS. I tried the same thing with the switch PS (300w) and the bulb lights up - though of course not very bright.

I have to set the switch PS to at least 18.5V to see any light at all (dark room). I can't get the Agilent that high.

Is this to be expected?

[newbrain]

My Agilent E3610A arrived yesterday and I have been trying Kleinstein's incandescent light bulb idea but haven't been able to light a 25w bulb.

Remember that this is a 30w PS. I tried the same thing with the switch PS (300w) and the bulb lights up - though of course not very bright.

I have to set the switch PS to at least 18.5V to see any light at all (dark room). I can't get the Agilent that high.

Is this to be expected?

Yes, if the bulb is rated for 120V...

[mos6502]

The reason I'm apprehensive about this unit is that it's incredibly light (a little over 4 lbs). Also, I've seen several teardowns of similarly priced "EBAY" power supplies and in every case they had much larger transformers than mine.

I'm not an expert and this is just a hypothesis, BUT, do you think that's maybe because yours is a switching power supply?

Also, like, what are you even trying to do?