Sh#t power supplies...

[mrflibble]

I don't know why you guys are hellbent on old crusty power supplies. For today's hobbyist, they suck.

Since you are in Europe, I agree 100% with you. Stupid crusty old power supplies. If I were you I would get a new Chinese one with nice digital readout.



That leaves more crusty old psu's on ebay for me.

[rdl]

You'll need mA resolution the moment you start designing things that have to run on batteries.

You are try to use the power supply as a meter, not what it's for.
For that you need a multimeter that reads microamps.

So in order to use one of those old supplies I need 2 DMMs attached to it on top of the ones I already need to check the circuit under test?

Edit: And yes, it should behave as a meter as well, that's why they are called bench supplies and not wall adapters.

No, you need a DMM to read micro amps, that's what they are for. Measuring. The power supplies only have indicators, they are for providing power, not measuring.

[sync]

As for the cheap agilent ones, they suck in two ways: 1) 10 mA reading resolution

Yes, that annoyed me. I have mod my E3610A to 1mA resolution. But how accurate are the current reading on the cheap chinese power supplies? Anyway, it doesn't matter when the turn on spike killed the project.

2) no way to set the current while you're outputting stuff. You have to push that damn button that basically shorts the output to ground.

Wrong. It doesn't shot the output. That would be silly. The power supply is working normally while pushing the button. It only switches the current meter from reading the actual current to the set current.

[rollatorwieltje]

I have yet to require mA resolution on a power supply, and I suspect this is similar to the people who think they need an 8 digit DMM.

Edit: The mA resolution (and lower) for output is useful for testing low power loads, like LEDs.

Bench power supplies are crap for that, most have a huge output capacitor and very slow current regulation. Basically it's only constant current to the output capacitor.

Also on my Agilent E3615a pressing the current set button certainly doesn't short the output, it only changes the input to the current meter.

[jadew]

As for the cheap agilent ones, they suck in two ways: 1) 10 mA reading resolution

Yes, that annoyed me. I have mod my E3610A to 1mA resolution. But how accurate are the current reading on the cheap chinese power supplies? Anyway, it doesn't matter when the turn on spike killed the project.

2) no way to set the current while you're outputting stuff. You have to push that damn button that basically shorts the output to ground.

Wrong. It doesn't shot the output. That would be silly. The power supply is working normally while pushing the button. It only switches the current meter from reading the actual current to the set current.

for 1). It's pretty easy to get really good mA reading and even uA.

for 2), if that's the case, I stand corrected, I was under the impression that it shorts the output in order to show the set value.

[UnknownVertigo]

As for the cheap agilent ones...

That being said, regardless of how simple linear supplies are and how easy it is to make the digital side, there seems to be a shortage of good bench supplies out there...

Thanks for all the good conversation folks! 

There was a question about my location.  I'm in the US...should add to my profile.

I liked the suggestion of using 2-3 e3610a supplies.  My issue is that as a bit if a noob I'm curious about the lack of OV/OC protection...especially if we're talking running in series or parallel.  This also begs the question about Dave's point on the supply not having a output on/off.  Seems like a supply better suited for the guys who know exactly what they are doing and can avoid damaging their circuit when hooking up multiply units that may not be aware of each other's existence.

Any thoughts in the next model up from the e3610a...something like the e3632a or e3642a?  More forgiving?

To the quote above.  Cheap is relative.  I've yet to feel like any test equipment is actually cheap.  Agilent always seems to be top buck.  Can you explain this statement more?

Also, you share a lot of information yet haven't commented on what your solution is to the issues mentioned.  What do you use?


Sent from my iPhone using Tapatalk

[jadew]

Bench power supplies are crap for that, most have a huge output capacitor and very slow current regulation. Basically it's only constant current to the output capacitor.

Yeah, that's definitely one of the decisions one has to make when making a bench PSU, good current regulation or good transient response. Personally I prefer super small output capacitance and have a capacitor (of the right value) in the project I'm powering, if I need it.

[sync]

for 1). It's pretty easy to get really good mA reading and even uA.

Ok, which power supply do you recommend?

[jadew]

@ UnknownVertigo and sync

I wouldn't know, I'm looking for a new supply as well and I don't know what to chose. I'd go for the Rigol but it's too big for my desk and I'm not sure I want to spend that much anyway.

Currently I'm using a home made PSU 0-20V, 0-1A with 5mV/1mA output resolution and 1mV/0.1mA reading resolution.

I'm contemplating building another one 0-30, 0-3A but I find it hard to find good enclosures and heatsinks with out paying a lot.

The heatsink could be shrinked by simply switching the transformer tap based on the output voltage (ignoring the set one), which is what I think the Korad one is doing and why it has that weird ramp up.

Anyway, I don't know what is best or what to suggest, but I strongly feel that old PSUs are not the solution for someone starting up with electronics or for the hobbyist.

[Frost]

Ok, which power supply do you recommend?

Agilent B2962A

[UnknownVertigo]

Ok, which power supply do you recommend?

Agilent B2962A

haha, awesome.  I'll buy two.

seriously though, supply...not source.  The difference between those two words is the difference between me sleeping in the bed or on the couch.

[sync]

I wouldn't know, I'm looking for a new supply as well and I don't know what to chose. I'd go for the Rigol but it's too big for my desk and I'm not sure I want to spend that much anyway.

Maybe this?
http://www.home.agilent.com/en/pd-1000001614%3Aepsg%3Apro-pn-66311B/mobile-communications-dc-source-15v-3a

Small, programmable and should be capable to precise accurate measure down to a few 10uA.

[Frost]

seriously though, supply...not source.

Ask as a person, who can't speak english very well, is there a big difference?
In the wikipedia I found: "A power source is an electrical power supply"

[UnknownVertigo]

Maybe this?
http://www.home.agilent.com/en/pd-1000001614%3Aepsg%3Apro-pn-66311B/mobile-communications-dc-source-15v-3a

Small, programmable and should be capable to precise accurate measure down to a few 10uA.

Specs look good but no terminals in the front...not the best for my bench.

btw, when looking for used units I know that pics can be telling.  What should I think when a pic shows a powered on unit with no leads attached but voltage and or current readings of anything but 0?

[sync]

btw, when looking for used units I know that pics can be telling.  What should I think when a pic shows a powered on unit with no leads attached but voltage and or current readings of anything but 0?

Depends. When it displays the set current then there is nothing wrong.

[UnknownVertigo]

Ask as a person, who can't speak english very well, is there a big difference?

You're right bud and your english is fine.  I read source and supply as the same.

Agilent seems to change their terminology though...and when calling a product a source it means many thousands of dollars.  If they call it a supply then it's more affordable.

Sorry for the confusion.  Just making light of that fact.




Sent from my iPhone using Tapatalk

[bingo600]

You'll need mA resolution the moment you start designing things that have to run on batteries.

You are try to use the power supply as a meter, not what it's for.
For that you need a multimeter that reads microamps.

So in order to use one of those old supplies I need 2 DMMs attached to it on top of the ones I already need to check the circuit under test?

Edit: And yes, it should behave as a meter as well, that's why they are called bench supplies and not wall adapters.

No, you need a DMM to read micro amps, that's what they are for. Measuring. The power supplies only have indicators, they are for providing power, not measuring.

Did you look at the specs. Of the hp6632b's i mentioned above. Their meter readout is more accurate than most handheld dmms. Mening you dont need to tie up one or more meters to read v/i , just look at the psu meters.


/Bingo

Suggestions

http://www.ebay.com/itm/HP-Agilent-E3630A-DC-Power-Supply-35-W-Triple-Output-6V-2-5A-20V/151229604982

http://www.ebay.com/itm/AGILENT-HP-6623A-TRIPLE-OUTPUT-SYSTEM-HP-IB-DC-POWER-SUPPLY-20V-50V-40W-80W/151229648884

http://www.ebay.com/itm/HP-Agilent-Model-6632B-0-20VDC-5A-Power-Supply-/171238502056

The 66xxoutput on the back isn't that Big a problem to Solve
 
The 66xxA models have LCD that can be Haarder to read than the VFD on the B's

Tooms has some info here
http://www.tooms.dk/?page=http%3A//www.tooms.dk/Tblog/Showblog1.asp%3FY%3D2012%26M%3D4

[Rigby]

http://www.ebay.com/itm/HP-6114A-Precision-Power-Supply-020V-2A-2040V-1A-65335399-/310873020499

[uoficowboy]

I don't know why you guys are hellbent on old crusty power supplies. For today's hobbyist, they suck.

Why you ask? Well, because projects are different than what they used to be 50 years ago. You need AT LEAST mA readings from the supply (and I mean digital display), otherwise you'll end up having to use a separate DMM for that. Same goes for voltage. Why have two extra pieces of gear just so you get something that a decent supply should already provide given today's technology state?

As for the cheap agilent ones, they suck in two ways: 1) 10 mA reading resolution and 2) no way to set the current while you're outputting stuff. You have to push that damn button that basically shorts the output to ground. That's... impractical to say the least and if you want 1 mA resolution you'll have to short them yourself, with your DMM (awesome! ... not).

Now... aside from mA reading, you need mA (edit: output) resolution! if you can go lower than that, that's even better!

Also, the bench supply needs to be programmable and have a way for you to output the readings to the PC so you can analyze and characterize stuff, as you advance (even as a hobbyist) this becomes a need more and more often.


That being said, regardless of how simple linear supplies are and how easy it is to make the digital side, there seems to be a shortage of good bench supplies out there, but I wouldn't go as far as recommending super old supplies or old designs that are simply unsuitable for what your usual noob or hobbyist does on his bench today.

My Power Designs TW5005W has a switch that toggles the full scale meter reading between 50 and 500ma. I can easily get single ma readings from that.

All the features you describe sound great, but are total overkill for 99.99% of what people do. You sound like you're describing an Agilent or Keithley source meter.

[jadew]

My Power Designs TW5005W has a switch that toggles the full scale meter reading between 50 and 500ma. I can easily get single ma readings from that.

All the features you describe sound great, but are total overkill for 99.99% of what people do. You sound like you're describing an Agilent or Keithley source meter.

I'm not crazy enough not to admit that in 99% of the cases you don't need that kind of functionality. That applies to me as well, however there's that 1% when you could benefit from such features and the fact that they're so cheap to implement makes me want a PSU that does that.

Yes, the cheap Chinese PSUs that have those features are not trustworthy because... well, bad design, not because you can't make a good PSU that cheap.

With the HP/Agilent ones you always have to make a compromise: price, chunkiness or specs. That is because they are overpriced (at least in the EU). Obviously good supplies otherwise, but that's a bit beside the point when all you had was $100-200 and the supply you got for that price doesn't do too much to begin with.

Finally, yes, I like the crusty supplies as well (heck, I might even get one in the end), but my point is that we're in 2014 and I expect to find a few supplies out there that are well built, have good specs (as the ones I listed) and are affordable. I just couldn't find that supply yet.

[Rigby]

For new supplies, pick any two of the following:

• Well built
• Well spec'd
• Well priced

The older the supply, the closer you can get to all three.

[echen1024]

For new supplies, pick any two of the following:

• Well built
• Well spec'd
• Well priced

The older the supply, the closer you can get to all three.

Exactly. Look at my HP/Agilent/Keysight 6114A for an example. Price: $60 shipped, around $100 for a good condition. Performance: Incredibly good, approaching that of a calibrator. Build: Old HP. Need I say more?

[uoficowboy]

For new supplies, pick any two of the following:

• Well built
• Well spec'd
• Well priced

The older the supply, the closer you can get to all three.

Exactly. Look at my HP/Agilent/Keysight 6114A for an example. Price: $60 shipped, around $100 for a good condition. Performance: Incredibly good, approaching that of a calibrator. Build: Old HP. Need I say more?

I will say that the current display of the 6114A is limited... I do wish it had a switch to select between different current ranges. Otherwise a really excellent supply.

Plus, if you have one of the newer versions, it doesn't have any sort of power on indication. Who thought that that was a good idea? Seriously...