the Poor Man 660 support thread

[analag]

Release the PSU schematic.

 

[jackies]

Release the PSU schematic.

Transmission from Mars detected!

 

[rrs]

If the PSU is giving you nothing but problems, why not build a new one? Breadboard a passive one or grab Tommypiper's extra board
and be done with it. I was thinking to go passive if the new PT doesn't fix my issues.  

We'll designing another PSU is way out of my league. I think either a modification to the existing one or if a PSU is already available from another project may be suited.

Hope some clever minds can nut this one out. It is well worth saving this project.

 

[emrr]

International Power linear regulated supplies.  Check Mouser. 

 

[tommypiper]

TommyP, what PSU are you planning on using? Care to share the design?

Not sure yet, except as simple and basic as possible.  

 

[rrs]

It would be simpler to just simplify the first regulator and forget any exotic parts.

Just use the design that is already there, the lower voltage regulator section has proven itself already. Do that with a 100-120V 3W zener and a trimmer in place of the 182K resistor. There's your new 250VDC regulator section, what, 5 parts?

This idea seems simple. I don't know how to do it myself but a diagram of what would need to be done might be a start.
Some opinions on why it would or would not work?

 

[rrs]

Still, those puny low rated and too small package transistors in a high voltage (and relatively high current) PSU is a basically a design error. They work fine in what we call a "happy day scenario" at work. Not a lot of you have had a happy day with this PSU. Start up is a place where bad things might happen, surge currents killing the little things.

This is exactly how my MPSA Transistors have blown in the past. They have to go.

 

[mitsos]

Release the PSU schematic.

Analag PSU Schematic

 

[kazper]

It would be simpler to just simplify the first regulator and forget any exotic parts.

Just use the design that is already there, the lower voltage regulator section has proven itself already. Do that with a 100-120V 3W zener and a trimmer in place of the 182K resistor. There's your new 250VDC regulator section, what, 5 parts?

If I'm reading you right your thinking you could just replicate the 136V supply but use an  100-120V Zener in place of D10?

This would sit in the same place as the 250V supply after C4?

 

[Kingston]

Yes, and have a 250-500k trimmer in place of the 182k to trim your voltage. Although sidechain B+ voltage is not terribly critical, as long as it's stable. Even as wide range as 200-270V is "close enough".

I've wondered what the meaning of all those terribly accurate resistor values is, 182k, 221k, 475r for example. None of these are critical places so just use the closest standard value instead.

Funny how time goes by, when I originally bought the kit I didn't think to question them. Now I just see "221K bleeder, what the... anything in the range of 100K-470K would work"

 

[analag]

Reduce the value of R1=10R, R8=22R, C9=22uF. Q4=IRF840, Q5=IRF820 or IRF840.Let's see why we're having issues.

 

[analag]

Still, those puny low rated and too small package transistors in a high voltage (and relatively high current) PSU is a basically a design error. They work fine in what we call a "happy day scenario" at work. Not a lot of you have had a happy day with this PSU. Start up is a place where bad things might happen, surge currents killing the little things.

So why am I not blowing up my puny transistors. I use this design in a few other things with no problem whatsoever. If you rated the transformer correctly you might not have deal with the ridiculous surge at start up.

 

[radiance]

deal with the ridiculous surge at start up.

If that's a problem than why not put a couple of NTC's in  front of the power toroid?
This way you can scale down the main fuse as well....

 

[mitsos]

forgot the heater schem for those that might want to see.. It goes trafo-->huge bridge rectifier--> Low R big W resistor, then to here.

(sorry for the quality, the scan is on my other PC so I took a pic with the laptop's built-in camera)

 

[Kingston]

If you rated the transformer correctly you might not have deal with the ridiculous surge at start up.

I had mine custom made by the best company we have around here. It still destroyed Q1-Q3 on very first initial testing when I was building the thing. I ended up basically p2p in that area because the PCB is quite low quality and does not take desoldering well (vias come right off). Worked fine ever since.

It crossed my mind there could be something overlooked in the PCB design that's killing them, but since it started working ok I could not be bothered to troubleshoot it further. Wider traces and less constrained layout would not hurt, and the PCB certainly doesn't (or shouldn't?) have to be double sided. Admittedly the PCB looks very pretty populated all symmetrical like that. Design goal priorisation problem? I would go simplicity first before making it look pretty in CAD, especially in the very error prone DIY kit field.

 

[emrr]

To repeat myself, if the PSU board should ever get a redesign, all the tall **** needs to go in the center, so the small **** isn't buried down in inaccessible canyons.  Put the small **** in accessible areas near the edges for ease of service. 

I have noted that the TO-220 devices do indeed tear the plating out of the vias when trying to remove one.  The holes are too small. 

 

[rrs]

Well the main problems I see are.
1) Why the small transistors are know to fail on start up.
2) Voltages dropping under load.

What makes troubleshooting so hard is the pads on the board will peel of on a hot day let alone a 300 degree soldering iron.

The compressor is great the SCamp and advice Kingston has added have taken things to a new level but for this to be a reliable unit I do thing a new approach to the PSU is needed. I understand the Mains Transformers could be of issue. I have the Avel Lindburg US group buy and am seeing 400V in places which does seem high and may be contribution to the problem but these things should be taken into consideration. Not everything will be perfect.

if we can get over this I do believe this compressor "PM660MK2" ?? could be one everyone will be proud of.

 

[kazper]

Mk2... If anything it would be a simpler and more robust PSU...

Knock on wood, no trouble here withe the mono....

 

[idylldon]

Knock on wood, no trouble here withe the mono....

No trouble here using the Avel Lindberg group buy toroid in a stereo unit and I've had mine running for 10-12 hours at a time.

Cheers,
--
Don

 

[radiance]

Knock on wood, no trouble here withe the mono....

No trouble here using the Avel Lindberg group buy  toroid in a stereo unit and I've had mine running for 10-12 hours at a time.

Cheers,
--
Don

These inconsistencies among people make it strange....