Instek GPD-3303D broken transformer winding

[richard.k]

First time posting, been lurking for a while.

I bought an Instek GPD-3303D power supply off of Ebay, "open box unused". It really does look and smell like a new unit, but unfortunately the seller didn't package it very well and it took some hits.

It appears to work but it's a little bent up, at the price I paid I might deal with it. But I found something weird when I was inspecting it for damage. The toroidal transformer has a spot where the outer winding was cut. To me it appears intentional, the wire was cut and it appears that one end was bent aside. I've tested the unit and it seems functional (every voltage range on every output, I know about the tapped windings). Is it possible that this winding cut was something intentionally done at the factory? The location of the cut is top center, under the handle brace. Nothing sharp there that could have cut into the windings.

Thanks! Richard

[SkyMaster]

I bought an Instek GPD-3303D power supply off of Ebay, "open box unused". It really does look and smell like a new unit, but unfortunately the seller didn't package it very well and it took some hits.

It appears to work but it's a little bent up, at the price I paid I might deal with it. But I found something weird when I was inspecting it for damage. The toroidal transformer has a spot where the outer winding was cut. To me it appears intentional, the wire was cut and it appears that one end was bent aside. I've tested the unit and it seems functional (every voltage range on every output, I know about the tapped windings). Is it possible that this winding cut was something intentionally done at the factory? The location of the cut is top center, under the handle brace. Nothing sharp there that could have cut into the windings.

Welcome to the forum richard.k,

The price you paid, was for the power supply before it was damage during shipping; because the seller did a poor job at packing it.

The damage to the transformer cannot be something done on purpose at the GW Instek factory.

eBay is supposed to protect you; you should contact the seller. If I am not mistaken, eBay suggest to open a dispute right away.

I suspect this is not the the reply you were expecting, but did you receive the power supply in the state you were expecting?

 

[richard.k]

Thanks for your reply! I understand about shipping damage and the Ebay return policy - unfortunately I've had to use it several times. I was just trying to weigh my options going forward. The thing is, the shipping damage is fairly minor and could be repaired, the unit appears to work properly - and the price was very good. My question was whether anyone had seen a factory mod like this. I have worked in electronics manufacturing in the past and I know that some manufacturers (including the one I worked for...) will ship things that aren't pretty from time to time. I don't have enough experience with Instek to know if this could have been their work.

[SkyMaster]

I have two GW Instek power supplies, a GPS-3030DD and a GPS-2303, both are well constructed.

Several persons on this forum also have the GPD-3303D, maybe some will share pictures of the internal of their own GPD-3303D.

 

[Armadillo]

First time posting, been lurking for a while.

I bought an Instek GPD-3303D power supply off of Ebay, "open box unused". It really does look and smell like a new unit, but unfortunately the seller didn't package it very well and it took some hits.

It appears to work but it's a little bent up, at the price I paid I might deal with it. But I found something weird when I was inspecting it for damage. The toroidal transformer has a spot where the outer winding was cut. To me it appears intentional, the wire was cut and it appears that one end was bent aside. I've tested the unit and it seems functional (every voltage range on every output, I know about the tapped windings). Is it possible that this winding cut was something intentionally done at the factory? The location of the cut is top center, under the handle brace. Nothing sharp there that could have cut into the windings.

Thanks! Richard

If the winding is cut, then only DC load testing can review the damage. I think maybe 1 of the range may not be working properly under load because there are internal relays switching inside. You will have to dc load test it from say 1v to 30 vdc at 3A to qualify the power supply. Otherwise you may count yourself lucky if its the 240vac primary side of the transformer which you are not using it, then better tape it up.

[Jono427]

I don't remember seeing this on mine when I got it a month ago but will try and check it tonight.  Can you take a picture of the front too?  If I remember correctly there are two different versions of this model and the spacing of the connectors is different.

[SeanB]

That is damage, and very likely to one of the secondary windings. the unit will work with one open winding, under low load, but one of the internal supply rails will not have a full wave rectifier on it, but only a half wave rectifier, and will have higher ripple on that particular rail.

not going to be a primary one, as those typically are well buried inside, but that is repairable by carefully slitting the sleeve further there and then carefully soldering in a jumper link, and then coating the area in a polyurethane varnish to insulate it again.

[richard.k]

Thanks, this is all useful information. I did check the primary windings with the input voltage selectors in all the positions and they appear to be intact. And I did check all the outputs under a pretty heavy load, but didn't look at ripple current. I'll do that, and look at the startup on my scope.

It appears to me that there is only one version of the 3303D, the current version still has the "gap" between the banana jacks.

The break would be easy to fix, as you say. I'm inclined to take a discount from the seller for the shipping damage and keep it around. My luck with Ebay shippers hasn't been great, I think about a quarter of the items I've purchased have had some sort of shipping damage.

[Kleinstein]

One could try to measure which output channel is connected to the broken winding. This makes it easier to locate a possible negative effect.  If the supply is still working the broken winding is likely on for the auxiliary supplies powering the control circuit. The control circuits don't need much power and thus could get away running on half wave rectification. It is only the channel with the display that needs a little more power.

If the broken winding is there for output current, the problem might not show up until a high output voltage is used, though no heavy load may be needed.

[edpalmer42]

Can you estimate the wire guage of the broken winding?

You might be able to tell the difference between an auxiliary winding that powers the control circuit and one of the main power output windings by the wire size.  The winding for the control circuit would have much smaller wire than the output winding.

Ed

[Jono427]

It appears to me that there is only one version of the 3303D, the current version still has the "gap" between the banana jacks.

Nope you are right - I have the GPD-3303S.  I think there might be two of those (but I looked at a lot of bench PSUs so I could still be thinking of another).  Looking at mine my transformer is a model "3003-PD003001" and it looks like yours did git hit really hard in shipping.  Unless your top support bar is much weaker than mine, it would take a lot of force to push it that far to cut the windings like that.

[james_s]

That certainly looks damaged, if they were going to disconnect a winding I can't think of many reasons it would be done like that. What I can't figure out is how that damage occurred, is there anything inside that the transformer could have smacked against?

[richard.k]

I spent the last couple of hours making small subtle bends to the chassis to get the front panel square, it looks pretty good now. The weight of the toroidal transformer bent the sheet metal, which is surprisingly soft.

I've tested all of the outputs under load, looked at the startup waveforms (well behaved on all outputs) and measured less than 0.5 mV of ripple on the outputs. I have yet to see if the USB works, but that's about the last thing left to test. I'm sort of at a loss about what that mystery winding does but I'll probably poke around until I figure it out. The cut wire is around 22-24 gauge. All of the connectors on the transformer windings have what seem to be reasonable voltages on them. I put a meter between the cut wire ends and saw 2.3 volts, which dropped to 2 volts when the variable outputs were loaded. I found a service manual for the 3303S online, I assume it's similar. That one has seven secondary windings, some with taps, so there's a lot going on.

The damaged spot in the winding is under the center of the handle brace, not the edges. The center is raised up a bit, there's no way the handle brace cut into the wire. Whatever happened, happened with the handle brace removed or the transformer out of the box entirely. Somebody did it on purpose.

All of the outputs run about 100 mV higher than the rated value, which is actually within spec for the box (the fixed outputs are only rated at +- 8%). Wondering if they did that on purpose to compensate for lead resistance.

After the discount from the seller for shipping damage I only paid around $90 for this unit. It still has some pretty strong "fresh paint" smell and the manual and cables are still in sealed plastic bags. I think it's worth a little extra effort.

[Decapitator]

With the power supply unplugged, I'd check for continuity from the cut ends of the broken winding to each of the leads brought out of the transformer. That would at least tell you which part of the circuit the cut winding "belonged" to.

[tablatronix]

In case of shipping damage check to see if your thermistors popped off the heatsinks there are green ones thermal glued on both sides.

[ciccio]

Toroidal tranformers are usually built with the primary windings inside (near the core) and secondary winding outside.
Thus the damaged winding should be a secondary.
Many times a secondary winding (which usually has an higher current rating) uses many smaller gauge conductors in parallel, instead of a single conductor of larger section.
This because it is easier to wind 4 four wires in parallel than a single, larger section wire. Other advantage: a better use of the winding area.

Maybe the damage resulted in one broken conductor in a multifilar wound secondary, and the power supply seems to operate correcly because it is not stressed to it's limits.
As others have suggested in the above posts, you can carefully try to repair the damage.
A suggestion: check carefullly the damage to look for a possible risk of a short, and eventually apply some insulation..
I've seen toroids with a similar damage (usually due to mishanlding before installation) develop a short with  time and heat cycling, and destroy themselves.
Best regards.

[floobydust]

It could be intentional- I have seen chinese incorrectly wind/tap transformers. To get (defective) product out the door, stuff like this is done. You need wire cutters to nick a winding sharp like that.

[SkyMaster]

It could be intentional- I have seen chinese incorrectly wind/tap transformers. To get (defective) product out the door, stuff like this is done. You need wire cutters to nick a winding sharp like that.

GW Instek is not an obscure Chinese company; GW Instek is Taiwanese. I do not know where the GPD-3303D is manufactured, but the GPD-3303D has a list price of $500.00 US; this is not a Chinese no-name bottom of the barrel quality power supply.

 

[floobydust]

It was purchased off eBay, so it could be a knock-off or clearing-house for factory duds.
Once the PSU is assembled and passed functional testing, who plays football with the power transformer?!

[richard.k]

Sorry I dropped this thread for a while when on vacation. The power supply is repaired. It turned out that there are two 8 VAC windings that feed the front panel board and the broken winding was one of them. The power supply was 100% functional even with the broken winding, which I would take as a sign of a conservative design (a good thing). I soldered it up and put some proper transformer varnish on the patch. And as I mentioned previously spent an hour or so making tiny little bends in the chassis to get everything to line up properly again.

I am pretty annoyed with the eBay seller. He asserted that it was "new" but the box was damaged. He actually packed parts of the original box with the power supply. Which included a label from Newark Electronics that says the power supply was received in June 2016 - and the condition was "damaged". I did get half of the selling price back from him based on the condition but I never would have paid the asking price in the first place if I had known he was selling damaged goods.

In any case it's good now. I think $90 plus some labor and aggravation is pretty good for a power supply that Newark currently lists for $415.