• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Welborne Terraplane Amp-PS10 HV Regulated PS Problems

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi, I'm having problems with my Welborne Terraplane Amp and it's PS10 HV Regulated Power Supply . Ron built these and the PS has already gone on one of the amps, which Ron kindly replaced. But now, I'm getting no response from Welborne and I've been trying for the last few months.

So having a look it at the PS10 it looked like their were burns marks around two of the diodes D7 & D8 and also the around the resistors R4 & R5. So I replace these but on powering up - spark and nothing as I powered off.

Parts used:
RSS 544-5436 Diode,Zener,200V,017AA,1N5388B
RSS 135-465 Resistor,carbon film,high stab,2W,5%,330K

Does anyone have any ideas on:
1. Fix this - it is a bit beyond me
2. Get a replacement unit
3. Remove these Power Supplies and replace then with something else as they have been problematic and not sure they are used to the 240+v we get in UK although Ron said they were.

Any ideas or suggestions much appreciated.
Russ
 
Were the amps working at some point? The 240V shouldn't matter as the PS won't see that unless the power transformer is hooked up wrong.

Post some pics of the boards if you can. I have one of these supplies, and I had the problems that the ground plane burned up in a spot -- my guess is that the copper was really thin, or impure, or something that would lead it to have a high resistance. This was back when Ron answered email and sent things that you paid for -- but he claimed he had never seen it before. I have my doubts as to the veracity of this assessment.

Anyhow, I ended up running buss wire over all the ground connections which worked fine. And, as I think of it, if the ground went in the same place as my ground went, it might send extra current through the zeners -- I think I replaced those at some point.

Also, do the parts names match the posted schematic? R5 bypasses the rectifier which seems unlikely to burn up. Is it a different part?
 
Hi dsavitsk, thank you for you reply. Yes both amps have worked, one still does with the replacement PS that Ron sent after the original failed. Unfortunately Ron didn't tell me what failed on the original, in various emails he suggested it could have been D1 and/or D2, the HV rectifiers or possibly even Q1 the Mosfet.

According to the PS10 manual PDF which is attached, I believe they are the resistors indictaed on page 3 as R5 and R6. Page two doesn't seem to tally with this showing "R5, R5"???

Also it show on page 2 it shows Z5, Z6 and Z7 and Z8 which I took to be D5 - D8. They were what looked like burn marks on the outside of the circuit board around where these components were attached to the board, i made a best guess at to what had failed based on that. Power Supplies like this are a bit beyond me.

Any help on how to fix this or how to test for which component has failed would be much appreciated. Thanks again for your help.
 

Attachments

  • ps10 manual.pdf
    74.4 KB · Views: 658
I thought you were saying R4 -- which is helps with the slow start stuff. If it is R5 and R6 that went then high voltage would seem to be a possibility. R5 and R6 are unnecessary, and you can simply leave them out if you like. Might they help the sound a little? Beats me, I've never heard it.

Yes, D5-D8 are the same as Z5-Z8. It does sound like the issue is component failure as opposed to PCB failure, and that high secondary voltage might be to blame.

Looking at the setup, it does seem that Ron had the transformer wired up the same for 220V/50Hz as for 240V/50Hz, which means that you might be getting a pretty high voltage on the secondary. Maybe the best thing to do is replace the rectifier with some overspeced parts

So, are D5 and D6 populated? Or just D7 and D8. The additive voltage total of these in series sets the regulator's output voltage. If they blow, it is because too much current is going through them. You set this with R2 -- there is a formula in the manual you attached, and if the voltage is too high, it will increase the current.

I'd do a couple of things -- first, look at how Ron repaired the other one and perhaps copy the values. Second, measure the voltage where R2 and C6 meet (on the one that works) so that you know approximately what the real secondary voltage is -- my guess is that it is a lot higher than it should be. Then use these values to replace R2 and Z5-8. It is also a good idea to just replace the mosfet as they seem to blow easily, and if it isn't working, everything else could go, too.

-d
 
Thnaks again dsavitsk, I will try some measurements of the working device when I get home and post later tonight or first thing tomorrow.


Sorry, yes it was R5 and R6 NOT R4.

D5, D6, D7, D8 are populated all the same I think.


I think the HV Rectifier Diodes are HFA06PB120. Can't seem to see them in the UK is there an equivalent. Same for Mosfet FQA6N80? I have PDF's for these if required.

I'm not sure Ron repaired it I think he just sent another one?

Hopefully talk more tomorrow.
 
Hi dsavitsk, apologies for the delay in replying, the usual weekend family crisis. I've managed to do some general measuring as listed in the manual.
TB3-3 to R5-1 is 590v - should be in range of 540-560
TB1-30 to TB1-12 is 153v should be 140-150
TB1-26 to TB1-9 is 5v and is OK

TB3-3 to HVIN on PS10 is 590v

I measured the R2 as 2.4K 3W - which is the same as the current one. The diodes I'd put in (two) were 5w IN 5388.

Voltage across R2 is 10v (I was not sure how to measure the secondary voltage - you mention to measure across R2/C6 is their any chance you could be a bit more specific about where to measure from?

Anything else I can measure?
 
First note -- you should use clip leads to take these measurements. Also, since you can pick up meters for $5 to $10, use Tubelab.com's multiple meter approach. 590V will put you flat on your back very quickly.

5MetersOFF.jpg


rjbaldwin said:
TB3-3 to R5-1 is 590v - should be in range of 540-560
TB1-30 to TB1-12 is 153v should be 140-150
TB1-26 to TB1-9 is 5v and is OK

TB3-3 to HVIN on PS10 is 590v

I measured the R2 as 2.4K 3W - which is the same as the current one. The diodes I'd put in (two) were 5w IN 5388.

Voltage across R2 is 10v (I was not sure how to measure the secondary voltage - you mention to measure across R2/C6 is their any chance you could be a bit more specific about where to measure from?

Anything else I can measure?

I am guessing that TB3-3 is ground? I'm not sure what the other test points are -- do you have a schematic labeled with them?

There is a bit of a discrepancy here. Two 200V zeners will set the B+ at 400V. If your high voltage is 590 and R2 is 2K4, this suggest about 13mA going through the Zeners (maybe enough to do some damage, especially if there is a surge or ventilation is bad.) On the other hand, 10V across R2 suggests 4mA going across the Zeners which is too low.

Also, do your amps really run with a B+ of 400V? Seems low to me -- by about 150V. But, I'd be doubtful that this PS can deliver 550V. Is the whole thing biased up by 150V somehow?

It also suggests that you are trying to dissipate ~15W from the Mosfet which is probably too much. I have a sense that Ron didn't really think this through, or that I am missing some information. Really only two 200V zeners?

I think if it were my amp, I'd pull the regulators -- use them for something else, and just build the unregulated PS from the original (which is a RCLC) with R=30R/5W, L=15H, C1=15uF, and C2=two 550uF 385V caps in series with a 330K/2W resistor parallel to each. The design of the amp keeps the PS out of the signal path for the most part, and you can parallel the electrolytics with some small film caps for a performance boost.
 
Thank you once again dsavitsk, your advice on the voltages does not go unheeded!
TB3-3 is a ground point, I'm not sure I have anything but the normal circuit diagram (on the web site) in pdf format.

Isn't it four Zenner diodes z(d)5 - z(d)8, I only replaced the ones where the circuit board looked burnt, so this could be a place to start. Or are your talking about the D1 - D2 rectifier diodes.

I suspect it is me not explaining it very well, I'm sure Ron has it nailed down, they sound very good when they are working. This is at the edge of my knowledge - I've never been very good at power supplies.
 
Okay, I just misunderstood. I thought there were only 2 zeners and they were both 200V which would have put it at 400V. (You could have jumpered the other 2.)

Did replacing the two zeners fix anything? Chances are no. I would think that if the two got hot enough to burn, any others in series are toasted, too.

So, I think what you should do is take some measurements in the working amp, and then we'll work from there. First, measure from the PS output to ground. Then, measure from the front of R2 (where it meets R1 and C6) to ground (the R1 and R2 on the PS PCB). This will tell us the input and output from the PS - I am guessing that there are probably a pair of 75V zeners in there, or something like it? It will also tell us the current across the zeners, though I think it is likely 4mA based on what you already said. This will tell you where the pS is operating, and whether you need to make some adjustments for long term reliability.

And, if you don't want to keep turning the tubes on and off, disconnect the PS from the amp and put a dummy load on it -- because it's a regulator, the load need not draw the same current as the amp -- just enough to bleed the caps so you don't zap yourself. Try a 10W to 20W 50K resistor.

The other thing to do is to simply replicate the parts from the first onto the second. Really, I'd replace everything except the caps. It will cost you about $10 and should solve the issue, even if it doesn't explain why it blew. It also might blow again someday ...

This is at the edge of my knowledge - I've never been very good at power supplies.

Me either, but this one is pretty easy. Basically, think of the Mosfet as being like a cathode follower that supplies current at the voltage set by the zeners ... This circuit might give you some insight: http://headwize.com/projects/showfile.php?file=szeke1_prj.htm
 
Sorry about the delay in reply.. As I had to be away for a few weeks, I dropped the amp over to Audiolabs in Leeds, who of course fixed it. As you rightly surmised dsavitsk, just a few components blown, although still not sure of the reason.

Many thanks for all you help dsavitsk. I probably could have fixed it myself, but lack of knowledge and time constraints were a deciding factor. If you are uncomfortable with what you are doing (especially with high voltages,) then perhaps you should not be doing it, was my view.
 
I have a PSU similar to the PS10 unit made be Welborne labs (mono blocks using teh Sophia Electric Princess 300B mesh with 2.5V filament). I think it's one of the previous models. Anyway, like other have found the PSU popped out while I was listening to music. I had a FET spare at hand and replaced it. Unfortunately I did not check any of the diodes and switched on only to find it popped the internal fuse with a loud bang.

Looks like two of the diodes have smoked, possibly the FET and the Zener across it too (not sure on that). The FET is a 2SK1511 which are hard to get hold of here in the UK (last time I ordered spares it cost me a small fortune) Anyone have any ideas of an alternative drop in replacement that is readily available? The diodes are labelled 8515 U860.

Ultimately, I'd like to do away with the regulated supply altogether as I think it's a design flaw. However, if there is some way to fix it either by substituting for better parts that can handle the load or by cooling it's something I'd like to do (don't have the requisite chokes on hand to build a CLC supply right now).

There is one big bummer though, I had a probe around and it appears the power transformer has a problem (it's a big electraprint transformer). The 2.5V AC winding has no power (replaced power fuse to probe) - something about the regulated supply going south has affected it (I may pull the HV windings from the regulater PCB and re-check later to confirm).

Any advice as to what you guys did to keep the regulated PSU working long-term would be appreciated..

Thanks!
 
Basically, you are looking for a TO-247 N channel mosfet that can take the voltage. If you look on Mouser there are a bunch that seem to fit the bill, ranging from $3 and up. Beyond that, the specs are not terribly important -- just check that the pin out is the same.

As for the PS, the design is not necessarilly horrible. It is just a zener to set voltage with a pass transistor. The problem is that the heatsinking is terrible, so finding a way to dissipate the heat better, or dropping less voltage across the reg, is your best bet.

It looks like you blew the rectifiers? U860 is a 600V part -- move to a 1200V part instead.
 
Thanks Dsavitsk, I'll dig around Mouser and see what I can find. Just a tad concerned about the AC 2.5V not firing at present (must be the fried FET and rectifiers causing that, so nee d to pull the AC from the regulator module and double check). Yeah those are rectifier diodes, I'll uprate those to something that can handle higher power.
 
Guys,

One hint.

The zenners pop due to overvoltage, most likely, the FET's possibly due to excess dissipation.

I suggest replacing the resistor with a TO-220 Current source, set so that worst case dissipation in the CCS part does not exceed around 0.5W.

Also, with either a variac or measured, run the amp at the lowest mains voltage you expect and increase R1 (or add resistance to the choke) to drop the voltage across the transistor as much as possible but stop when you can measure the AC noise on the output of the supply rising significantly.

This way you remove heat from the series pass transistor (which easily breaks) and put it into the resistors (which can handle excessive power a lot more gracefully. Obviously, use high power resistors...

Normally it is easy to calculate the needed resistance by checking how much current flows (I = V/R where V is the voltage across R1 or the Choke and R is the value of R1 or the DCR of the choke) and checking how much extra voltage is needed to be dropped, then simply R = V/I.

Ciao T
 
Folks,

I suggest replacing the resistor with a TO-220 Current source

That is replace R2. Also, set the current so you do not exceed the Zenner diodes power rating.

It is important to realise that 5W rated zenner diodes only can handle around 1.3W at 25 Centigrade unless substantially heatsinked. So if we allow for hot temperatures (as found in a Tube Amp) a 5W zenner diode should not dissipate more than around 0.7W or it will be prone to poping (see datasheets).

For a 200V/5W Zenner Diode one should hence keep the current to around 3.5mA which is much less than what is suggested in the manual.

If I wanted 550V from such a supply I'd probably use a bunch of 150V zenner diodes (5W type) at a low current so they produce only around 140V actual voltage (see curves in datasheet for the exact diodes.

Anyway, just some more quick notes after reading the manual.

Ciao T
 
Hi,

I think I must've popped the Zeners too. I have replaced the rectifier diodes and the FET in both amps and neither of them are working.

Just measure the Zenner diodes using a multimeter set to "Diode Test", one direction they should measure "open circuit", the other way around 0.6..0.7V. If not they have gone and joined the Norwegian Blue Parrot.

Ciao T
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.