the Poor Man 660 support thread

[matta]

Hi Guys,

I'm finally getting around to building my PM660!

It's just a single channel version, using Moby's DC heater board. Wired her up and flicked the switch, but no heaters under load  :-[

I checked my PSU's before connecting and wiring the PSU to the audio board and adjusted accordingly, all good.

It's a custom wound torrid and I beefed up the windings, 9A on the 9V winding, which is giving me 11VAC unloaded, going into Moby's board I've got 11.7VDC from the bridge rectifier, trimmed it to 6.3V as specified, but the moment I connect the heater out of Moby's board to the heater in on the audio PCB I get 0.00V!

Any ideas on where to look? Could it be a bad regulator? Looking at the schematic KL1 and KL2, which are the heater headers only connect to the heaters, nothing else, so not sure what could cause this?

Any help would be much appreciated!

Cheers

Matt

 

[Moby]

Hi Guys,

I'm finally getting around to building my PM660!

It's just a single channel version, using Moby's DC heater board. Wired her up and flicked the switch, but no heaters under load  :-[

I checked my PSU's before connecting and wiring the PSU to the audio board and adjusted accordingly, all good.

Underload my voltages are as follows:

139V instead of 136V
260V instead of 245V
-17v trimmed to -17V
+17V trimmed to +17V

It's a custom wound torrid and I beefed up the windings, 9A on the 9V winding, which is giving me 11VAC unloaded, going into Moby's board I've got 11.7VDC from the bridge rectifier, trimmed it to 6.3V as specified, but the moment I connect the heater out of Moby's board to the heater in on the audio PCB I get 0.00V!

Any ideas on where to look? Could it be a bad regulator? Looking at the schematic KL1 and KL2, which are the heater headers only connect to the heaters, nothing else, so not sure what could cause this? Bad tubes?

Any help would be much appreciated!

Cheers

Matt

Hi Matt, this psu design is so simple so besides wrong caps polarity U can't make other serious mistake. I think that you have same issue like me (fake regulator) check here http://groupdiy.com/index.php?topic=28274.msg435997;topicseen#msg435997. Cheers

 

[matta]

Hi Matt, this psu design is so simple so besides wrong caps polarity U can't make other serious mistake. I think that you have same issue like me (fake regulator) check here http://groupdiy.com/index.php?topic=28274.msg435997;topicseen#msg435997. Cheers

Hi Moby,

Ah, I clearly missed that! Fakes? Balls! Well I'll try a different supplier and see if I have better luck. As you said, it's pretty straight forward, the caps and diode are all orientated correctly, hence my guess at it being a bad regulator.

Cheers

Matt

 

[matta]

Right, new LM883T's acquired, different supplier, batch code and looks to be authentic.

Unloaded I can trim it just fine, but waiting to work out what to do about the audio trimmers RV3 and RV6.

Volker said this in an earlier thread.

RV3 is adjusted to -2.4V and RV6 is set to -4.5V between Pin2 and ground.

Should this be done without tubes or with the tubes in place? I'm also not sure I understand where to do it exactly, is that Pin2 on the tube socket, or pin 2 on the trimmer, which in the case of RV3 is feeding the base of Q1, the 2N6107.

Likewise with the RV6?

Thanks in advance guys, help is as always, much appreciated.

Matt

 

[duantro]

Hey Matt,
I soldered a 2" test wire to pin 2 of the trimmers, under the pcb, that is sealed in shrink tubing, unless I need to adjust voltages. Then just check this pin 2 lead, with a dmm to ground, while the tubes are in.
Hope this helps!

 

[matta]

Thanks man, that's a fine suggestion!

Hey Matt,
I soldered a 2" test wire to pin 2 of the trimmers, under the pcb, that is sealed in shrink tubing, unless I need to adjust voltages. Then just check this pin 2 lead, with a dmm to ground, while the tubes are in.
Hope this helps!

 

[matta]

Hi Guys,

Ok, some success, got the heater board working, tube light up, nothing smoked, seemingly ok.

Tried to pass audio and it sounds like someone farting and hum.

Rechecked the voltages and noticed that with the tubes in the 139V rail drops to just 42V and the 260V rail drops to 20V  :-\

Any ideas? I don't see any damage, transformers are all connected, triple checked the off board wires to the unit, all seem good.

If I pull the tubes the voltages jump back up to to their regulated voltages, well a little shy, getting 220V on the 245V rail and 117V on the 139V rail.

AC voltages are good from the transformer, even under load.

As per duantro's tips I soldered and heat shrunk a test prove on the RV3 and RV6 and set the bias accordingly.

Any thoughts or suggestions are, as always, much appreciated!

Cheers

Matt

 

[matta]

Following up on my previous post a few days ago. As mentioned, the voltages dropped fast when one or several tubes were inserted.  So now, finally got my voltages right, also with tubes.  i've been looking for shorts on the audio-pcbs for weeks, but it must have been something with the voltage-regulators.  I changed them from two IRF840A into two IRF840B !  Looking at the voltage-regulators spec-sheets, I don't see why this should make much difference, but it does certainly work now. Another option might be that I had too much thermal compound, and that the heat didn't transfer fast enough.  Anyway, it could also be the 840B that made the difference. So all you guys struggling with voltage-drops. Take a look at the those regulators again.

So, I TRIPLE checked my wiring and parts values, all look good. I have a single channel unit and back in the day IRF820's were suggested for both Q4 & Q5, though at least 3 member seemed to have had the similar or same issues which were resolved by substituting both the IRF820's for IRF840's, actually the B variant, though these aren't readily available here in South Africa.

I'll try track down some IRF840's on Monday and substitute them and hold thumbs that it resolves by voltage issues  :-\

EDIT:

So I noticed this was posted as well.

PSU issues:
C6 is a Electrolytic and should be a Film 1uf Max, but 0.47 Uf is best... so C6 = 0.47 uf 350-400V Film cap.

Mine is stuffed with the original screened part, the 47uF electrocap, could this be the issue I'm having with the voltage drops?

Cheers

Matt

 

[duantro]

I was about to ask if you saw some of the power supply mods, like the film cap suggestion that you mentioned.
I go a temperature controlled thermistor fan installed in the top of my rack enclosure. Nice and quiet also!

 

[alexc]

IRF840s are a must imho.

For now, power up only 1 channel with HV - on the other channel, leave the heaters and bias voltages but disconnect both HV wires - insulate the ends and leave them for now.

Concentrate on one channel only at a time until you get stable.

On the channel under test, pull both 5687s and just look at the 6BC8s for now.

The PSU is fine with all this - just make sure you have the traffos with HV T2 and T4 fully connected at all times.

With no 5687s at all in circuit, the +245V should be mostly idle and able to maintain voltage. Make sure it does.

That means the +136V won't be also dragged down - so check if it can maintain voltage with the 2 6BC8s in circuit.

That will narrow down the possibilities for excessive loading on the HV - could be the pcb or the each of the pairs of tubes doing the thing.

Apart from possible bad tubes or solder bridges on the pcbs, the most likely things are bad psu or bad T2, T4. imho xsm/wsm do have a tendancy to fail, either on delivery or in initial use.

I hope you checked the continuity on delivery - a must QC check for these hit-n-miss units. There is a significant doa rate, imho. I'm pretty sure I fritzed one or two in initial use too. Once they are working past 'burn-in' they seem to be fine, though.

PS - make sure the CT with HV at T2, T4 is well insulated at the solder post - they can be lethal.

Also, with the psu the sub of C6 as discussed earlier in this thread can be made, but I don't think it makes a difference.

If you have cause to repair the psu board, a couple of psu changes as described by 'ppa' can be beneficial.  ie. there's a couple of protection diodes that make a difference.

I wouldn't do it as a matter of course for now - see how you go. I did it in 1 of 4 psu boards only.

DO check the insulation of the IRFs to their heatsinks is working - they can go to HV if not properly insulated.

Cheers

 

[duantro]

I concur with Alexc on checking the transformers. I had a handful of them dead, and die randomly on me, causing me lots of confusion while trouble shooting.

 

[alexc]

And I would love to try a poorman with Cinemags. I hear they can be great there

In fact, if I was  a few years younger I'd love to do another one and really go for mastering quality!

 

[duantro]

And I would love to try a poorman with Cinemags. I hear they can be great there

In fact, if I was  a few years younger I'd love to do another one and really go for mastering quality!

I really, really, dig the cinemags. Super smooth sounding!

 

[matta]

Hi Alex,

Thank you SO MUCH for jumping in and offering your thoughts and suggestions on the matter.

I've tripled checked for bridges, wiring wrong parts etc, though I'll admit that I just took the transformers as 'working' and didn't ohm them out before use. I assume one just measures the primary and secondary to check the DC resistance? Were there ever readings given as to what theses should be, even ballpark?

I'm only building a single channel unit, so only have one PSU and Audio board, but as per your suggestion I pulled the 5687's and found that there was still a drop, about half of the unloaded voltages.

136V = 77V and 245V = 146V, thought with the all 4 tubes I was getting 42V on the 136V and 20V on the 245V.

I don't have another pair of tube to check with, but a single tube and swapping the single single tube with one from the pair that is in there gains me and extra 10V on the 136V rail and 20V on the 245V rail which jumps to 167V, but still a considerable drop.

The current IRF's are well insulated, with Mica insulators, both are working, keeping that HV off the heatsink.

As shared, currently I have a pair of IRF820's in there, on Monday I can get a pair of IRF840's and swap them out to see if it makes any difference, though I suspect before I go down that route I should check those audio transformers!!!

If you have any other thoughts or suggestions they would be much appreciated, thanks for your time.

Duantro, thanks for the tip, yes, I caught the cap tip, noticed that Alex incorporated it in one of his units, though, as he shared, I suspect I should elsewhere first before going to the cap for answers.

Cheers

Matt

IRF840s are a must imho.

For now, power up only 1 channel with HV - on the other channel, leave the heaters and bias voltages but disconnect both HV wires - insulate the ends and leave them for now.

Concentrate on one channel only at a time until you get stable.

On the channel under test, pull both 5687s and just look at the 6BC8s for now.

The PSU is fine with all this - just make sure you have the traffos with HV T2 and T4 fully connected at all times.

With no 5687s at all in circuit, the +245V should be mostly idle and able to maintain voltage. Make sure it does.

That means the +136V won't be also dragged down - so check if it can maintain voltage with the 2 6BC8s in circuit.

That will narrow down the possibilities for excessive loading on the HV - could be the pcb or the each of the pairs of tubes doing the thing.

Apart from possible bad tubes or solder bridges on the pcbs, the most likely things are bad psu or bad T2, T4. imho xsm/wsm do have a tendancy to fail, either on delivery or in initial use.

I hope you checked the continuity on delivery - a must QC check for these hit-n-miss units. There is a significant doa rate, imho. I'm pretty sure I fritzed one or two in initial use too. Once they are working past 'burn-in' they seem to be fine, though.

PS - make sure the CT with HV at T2, T4 is well insulated at the solder post - they can be lethal.

Also, with the psu the sub of C6 as discussed earlier in this thread can be made, but I don't think it makes a difference.

If you have cause to repair the psu board, a couple of psu changes as described by 'ppa' can be beneficial.  ie. there's a couple of protection diodes that make a difference.

I wouldn't do it as a matter of course for now - see how you go. I did it in 1 of 4 psu boards only.

DO check the insulation of the IRFs to their heatsinks is working - they can go to HV if not properly insulated.

Cheers

 

[alexc]

No probs Matta - just hope you can persevere and get this guy working

For one channel only, IRF820 should be fine.

Checking traffo for continuity means, when not installed - put the multimeter to low ohms or 'beeping' resistance/continuity setting and put the probes briefly on the ends of the coil.
Look for low-ish resistance or 'beep'.

In the case of an uninstalled xsm/wsm primary of 600ohm specified coil would be around 50ohms dc or so. Secondary of specified 10K coil would be around 900ohms or so.

If the traffo is installed already, you can either i) unsolder the connections and do the above (POWER OFF!)

or  for the case of T2 primary (B+ power applied to center tap) - expose the solder taps of the ends of primary and carefully look for B+ (136V) at each end of the primary with respect to ground.

ie. red probe to one end, black probe to ground, 200V range.

If using a CRO :

DONT put a CRO (internally earthed probe ground) across T2, T4 primary with HV present.
You'll short HV to ground thru the CRO. Big bang!
Do it probe tip to HV hot and probe ground to .. ground.

If there is continuity, the HV will appear at each end (less a couple of volts for the coil losses)

For T2 secondary, just do the dc resistance check, powered up or down - no dangerous voltages there.

Same T4 really, but 245V there.

T1 and T3 primary and secondary are both low voltage - no danger there.

Basically you just want to make sure they are not open circuit (or partial short circuit!) - those fine little wires are exposed at the terminals are easily broken (open circuit condition) or can be with insulation faults (partial or short circuit).

---

Since you still have no joy, pull the 6BC8 and again check for HV rails. I think you said no tubes = good HV so if that's the case then it would appear 5687s are dragging down 1/2 the HV and the 6BC8 the rest 

No est bueno.

So - again, verify no tubes = good HV, verify heaters remain at 6.3V with/without the 6BC8 tubes and then stop.

Check again your wiring - the signal transformers first.  Then check the attenuators too.

And let us know again what you find.

---

Bottom line is this - if the open circuit HV is correct and adding tubes causes it to droop dramatically, then you must be pulling too much current somewhere. It can be either the tubes (unlikely) or the pcb traces shorting/bad component values (unlikely)  or the HV at transformer T2, T4 primaries are incorrectly connected/faulty OR the PSU is bad (most likely)

The PSU can be bad with incorrect component value or shorting traces or bogus part (not generally likely).
I find if the unloaded PSU HV is OK, then the PSU is generally OK.
 

---

There's not much more there, with respect to HV

- PSU 136V goes to the CT of T2 primary and to the tubes. That's it.
- PSU 245V goes to the CT of T4 primary and to the tubes. Again, that's it.
- The 136V is sourced from the 245V , so if the higher is nfg, so too will be the lower.

If they are open circuit, (or no heater) nothing works but they maintain their voltage.
If they are partial short circuit, they droop.
If the PSU can't take the current, it burns up.

Good luck - mostly worth it when it works

 

[alexc]

Most of the time, the HV issues are bad wiring or nfg T2, T4.

Bad PSU usually means smoke or bang or both.

Audio problems are usually bad heater or bad wiring of attenuators or the HV issues above.

 

[matta]

Hi Alex,

Thank you for your definitive post! I've just taken the unit apart and measure the transformers, I had a bad T3 in that the secondary was open, on closer inspection the wire had snapped off the terminal, thankfully I was able to get to the loose wire and solder a new piece to from it to the terminal, now all seems good there, below are my resistances, which seem to be in spec with the data sheets I pulled from Edcor's site for the XSM transformers I'm using.

T3
1+2 = 55R
1+4 = 118R
5+8 = 260R
6+8 = 502R

T1
1+2 = 55R
1+4 = 118R
5+8 = 262R
6+8 = 497R

T4
1+2 = 220R
1+4 = 463R
5+8 = 63R
6+8 = 120R


T2
1+2 = 220R
1+4 = 463R
5+8 = 63R
6+8 = 120R

I was hoping that may have resolved the issue, but alas the problem persists.

As per your recommendation I pulled all the tubes and measured the PSU voltages without a load, they are as follows:

HV AC IN = 265VAC
LV AC = 32VAC

Unloaded
Heater = 6.3V
245V = 222V
136V = 118V
-17V = -17V
+17V = 12V

Loaded with just the 6BC8's
Heater = 6.3V
245V = 143V
136V = 75V
-17V = -17V
+17V = 12V


Loaded with just all valves
Heater = 6.3V
245V = 40V
136V = 20V
-17V = -17V
+17V = 12V

B+ on T2 between ground and Primaries is 64V with just the 6BC8's.

There seems to be no visible damage on either the audio or power PCB's, no smoke, no burnt parts, though it's no guarantee, but I guess it's still a good sign?

Any thoughts from here?

Thanks again for all your help!

Cheers

Matt

 

[alexc]

Hi Matta

Just doing a quick measure on an xsm 600:10K by way of example.

(primary 600) pins 1-2  53 ohms
                            2-4  59 ohms
                            1-4  112 ohms

(secondary 10K)    5-6  250 ohms
                            6-8  230 ohms
                            5-8  480 ohms


These things can vary a bit, depending on how long you measure for, excitation power etc.
I used the approx 10% rule-of thumb before, but that's what I get right now (more like 5%)

So your T1, T3 look OK, and so do your T2, T4

Good.

Unloaded voltages :

-> AC HV looks OK -> indicates +344Vdc HV B+ before dropping resistor
-> AC LV looks OK providing the '32VAC' you mention is in fact 16V-0-16VAC center tapped

-> DC HV unloaded are too low
-> DC +12V fine too low as long as you can adjust. Sometimes you need to turn the trimmer all the way to each end to 'engage'

Need to sort that before going on.

I'll do some more tomorrow - sleep now

No smoke is a good sign! These guys will burn up quick if not happy and stay cool if no reason to misbehave!
Regardless of anything else - the only time I've had the HV go is when I've shorted it, wire-jagging carelessly.
It makes a good light, sound and smell show. For a moment.

PS - when powering up tube supplies, and until stable - good idea to use eye protection
      (and ear protection if you are like me .. a little .. twitchy)

PPS - use an insulated screw driver for trimmer adjust - it's easy to 'fly off the handle' and land on a HV spot.
        no fun with an uninsulated jewelers driver!

PPPS - make sure you have a 'hv drain probe'. If not, you can seriously shorten your life expectancy with tube builds.
          you need to 'make safe' an open chassis after each power-down WITHOUT EXCEPTION - TEST EXPLICITLY the HV

I have had the trimmers go - mostly cause I used cheapy bay types. Best spend good for trimmers.

Cheers

 

[alexc]

OK - even with anaemic HV B+ at the first regulator (+245V), your 6BC8 in-place are causing a 50% drop in the +245V rail.

The 6BC8s dual stages take only 3-6mA a-piece so you have either a nfg HV supply or your 6BC8s are current pigs (not likely).

Given the mostly monstrous HV current capability of these PSU traffos, I conclude your PSU HV section is NFG.

No matter - just step by step thru the mine field.



Without further checks, I would say

i) wrong value dropper  (possible but not likely)
ii) fake reg  (not likely)
iii) toasted little transistors  (likely!)

There's a couple of other bits in that 'heavily populated and mostly not needed' middle section that can be an issue.
(god help us! (us!))

You'll be needing some 'solder wick', a steady hand and some flux-remover
AND a tooth brush.

Cheers

 

[matta]

Thanks Alex!

Good news on the transformers.

Yes, the DC HV unloaded does indeed not look all that good? Any thoughts on where to start looking, probing to see if I can weed out potential issues with the transistors? Not looking forward to desoldering those DS boards  :-\

RE the DC +12V, I can of course get higher, it's trimmed down to +12VDC because I'm using 12V relays for the bypass board.

I've been using an insulated screwdriver for the exact reason you mentioned, WAY to easy to slip in there around those HV's!

I've checked and rechecked all the resistor values, I ALWAYS measure before stuffing boards, each resistor value, takes longer but prevents issues down the line, did the same here, but just to be sure I printed off a copy of the PCB PDF's and checked and rechecked each value based on the colour bands. All good.

Thanks in advance, I appreciated your patience and support, I'm confident we'll get to the bottom of this.

Cheers

Matt