power supply "sticks" in current limiting mode

Well, here's a schematic at least...

-- Adam

I'm not sure I understand exactly how your supply works but my 25 year old analog B&K power supply would shut down and indicate an over current condition when there was none. It seems I was always pressing the reset button. After some careful troubleshooting, I found that the front panel current limiting control pot was intermittent. I cleaned it and now it as good as new.

HTH.

--
David Farber
David Farber\'s Service Center
L.A., CA

That's called foldback current limiting; it's a positive feature, but not in this kind of power supply; I suspect something is broken. Open the box up; if there's a schematic printed inside, you're in luck. Otherwise, maybe there's a switch labeled 'foldback current limiting'?

Well I see one annoying thing.. whoever made mine swapped the digits around in the model number to make it "unique." This one looks close enough anyways. Now if I can just figure out what's going on in there.

Given that two lefthand pots set the current-limit, have you tried turning them up?

--
    W
  . | ,. w ,   "Some people are alive only because
   \\|/  \\|/     it is illegal to kill them."    Perna condita delenda est
---^----^---------------------------------------------------------------

I have the exact problem with my HY300D-3.. Have You found a solution ?!

Mine puts out full current, but has the unusable current limit latchup. Almost impossible to charge a battery with it.

Reverse engineering it exceeded my attention span, given that the schematic doesn't match the unit.

Whoever designed the thing must have been standing on his head and drunk. I stuck mine under the bench. Decided to redesign it and throw out 3/4 of the parts if I ever get really, really bored.

Holler if you find the solution.

Yep, you can read dates. Doesn't change the fact that at least two of us still seek a solution.

a solution. "

True, but realize that some newsreaders have this thread waaaaaaay down at the bottom of a very long list. I see you use gmail, so do I. When you post a reply gmail brings it up to the top. This is not always so with people w ho use other methods of access. To get to the maximum number of people it i s better to start a new thread, after seeing if any of the links are good i n the old one of course.

It is a bit surprising that there is still a print available via that link. If you see an old post like this and a link is to a tinypic or an imagesha ck, it is not likely to still be there. Russian (.ru) links seem to stay pu t, possibly because they are not a throwaway society and keep things runnin g, instead of dumping everything in a landfill every two years.

Anyway, according to the print, current sensing is done by R15. Anythng wro ng with N2 or its associated circuitry could be at fault.

Question, what does the current meter read ? Is it zero or does it read a b unch of current ? It could be as simple as something shorted across the out put terminals.

I assume the constant current LED is lit, if you turn the voltage all the w ay down does it switch to the constant voltage LED ?

First problem is that there seems to be a bunch of similar designs with the same model number. Mine doesn't match this schematic.

Second problem is that it's an upside-down backward design. Negative supply floating with the most positive end the internal ground reference. Theoretically, not a problem. Practically, it makes it harder to instrument/diagnose...especially without a correct schematic. The digital readout is mixed up in all that floating stuff and the schematic for that part doesn't seem to exist anywhere.

This appears to have been "designed" by someone who had never used a bench power supply, much less ever designed one. You can do a working power supply with far fewer parts. It's a crap design with unbounded op-amp inputs that can go "out of bounds" and latch up...patched with a bunch of zeners to try to limit that. But there can be and apparently are transient issues that can latch it up. Another consequence of this is that you can't set the current limit near zero. I had to adjust the minimum to 340ma to keep it from latching up.

While I'm on a rant, I don't like the way the thing works. The fine voltage control is a percentage of the coarse setting. So, with the coarse at zero, the range on the fine setting is from zero to zero. The whole purpose of digital readout is to know the outputs.

0.1V resolution is insufficient. The power switch doesn't turn off the output. Output stays up for a long time after you cut the input power. Who knows where the transient voltage/current go during that time.

Mine seems to work kinda normally as long as you don't need a current limit lower than 340ma and you don't have transient load conditions near the limit setting and you don't try to hook it to something with voltage, like to charge a battery. The only things that don't work are the things I most expect from a bench power supply.

My conclusion was that it isn't really broke; it's a crap design doing the best it can...and wasn't worth redesigning. But if someone comes up with a quick patch, I'd do it.

My response to the posting was, "me too", "let us know if you find a solution." Ranking optimization was not a priority.

Actually I didn't want to pontificate but I was going to say that it looks like someone studied instrumentation design but forgot basic circuit design . Someone should tell him that more components don't make the design better .

I think using the upside down design as you put it results in a bit better stability, that is if you know how to do it. If the thing can't foldback lo wer than 340mA, they didn't know how to do it. In that state you couldn't t est a 12 volt 1 watt Zener.

If the design is anything at all like the one posted, I don't see any quick fixes. I could probably redesign it faster. Maybe you should guve that a g o. You have a transformer and a box, and apparently two readouts.

I think some of these kits might just be junk. Not too long ago I was asked to look at an old Heath/Bell & Howell scope. It had an intermittently shor ted HV cap and a couple of minor porblems, not the least of which is that i t had never been aligned. This ain't the first time...

Anyway, it is DC coupled but the sweep generatior doesn't have a split sup ply, so when you go from external to triggered to freerun the trace needs t o be recentered. At first I couldn't believe it, I thought there was someth ing wrong with it buit nope. That is how it works. I've seen them only use short little ramps at higher seep speeds, which I really don't like, but no t this before.

Maybe that's where some crappy designs go when they got too much money into them - kits. Then the buyilder thinks he screwed it up and everything is f ine. The only problem is if they give the option to send it in for repair, but then someone else gets a crack at it. Business is business.

Seems like you could keep their power transistors and big resistors, take a coupe of your own OP AMPs and make it into something useful.

In the meantime you might want to repost, or I think they call it "top post " this so everyone sees it. SOmeone might have an idea who uses Thunderbird or something to get here. Alot of the oldtimers do.

design but forgot basic circuit design. Someone should tell him that more components don't make the design better.

how to do it. Not sure that's true. I'd bet that someone did it because they had NPN transistors and didn't know any better. Then that design got stolen and "improved" by some competitor...repeat a few times...add digital readouts...and this is what you get. This does not look at all like a design. Looks like a bad start that kept getting patched instead of fixing the design.

If the thing can't foldback lower than 340mA, they didn't know how to do it. In that state you

couldn't test a 12 volt 1 watt Zener. It's not foldback. There are pots galore. You can set the low limit below 340 ma, but mine latches up if you do.

it faster. Maybe you should guve that a go. You have a transformer and a box, and apparently two readouts.

snip

s like you could keep their power transistors and big resistors, take a coupe of your own OP AMPs and make it into something useful.

If I wanted another analog supply, I've got a design I did over 40 years ago that works fine. As you probably know, the schematic is trivial. The "art" needed for proper operation is a scarce commodity. I've fixed many "production-ready" designs by experienced engineers. What shows up as an element on the schematic is only a part of what's important.

Then, there are issues with making the meters autorange and the relays that switch the transformer...and the circuit board and and and and... It's all a royal pain in the ass.

I don't need another power supply or another project.

SOmeone might have an idea who uses Thunderbird or something to get here. Alot of the oldtimers do.

>