Pots at B+ Voltage

[Steve Conner]

Hmmm......

My experience is that, sadly, not every manufacturer thinks like this.

Well, this is the Theory & Design board. It's not about how amps are, it's about how they ought to be.

[guitarmike2107]

Sure it works, It's simply piss.

That sounds like loads of guitar amps, valves run way past the recommended plate voltages, biased hot etc etc
so it fits in well

Merlin has added some high value resistors in his circuit in case of pot failures, or the wiper loosing reference.

How is it bad "engineering" using a part that is specified for a maximum voltage and power rating higher than the application. I am a mech engineer.. this just does not compute for me.

I can see the value of using the more complicated circuits in a commercial amp as you can then use dirt cheap pots and save allot more than shelling out for expensive PEC pots.

[Steve Conner]

I don't think I've ever seen a pot rated at more than about 350V element voltage. Mil-spec or 2 watt doesn't necessarily imply high voltage rating.

Obviously if you could get one, it wouldn't be bad engineering.

[defaced]

How is it bad "engineering" using a part that is specified for a maximum voltage and power rating higher than the application. I am a mech engineer.. this just does not compute for me.

It's not about data sheets, it's about best practices. As I'm sure you know, best practice does not abide by data sheets or any published literature, it's about objectively looking at the design and thinking "does this make sense". I'm a welding engineer, and pretty much everyday the text book gets left in the office while I play "best practices" making test welds.

Or to put it simply, my amp (or really test rig in this case), my rules. I've felt 500v, I don't ever want to again. And if that takes a more complicated solution that one already out in the wild, then I'll develop it with the help of people who know what they're doing.

[guitarmike2107]

I don't think I've ever seen a pot rated at more than about 350V element voltage. Mil-spec or 2 watt doesn't necessarily imply high voltage rating.

Obviously if you could get one, it wouldn't be bad engineering.

heres the data sheet
http://media.digikey.com/pdf/Data%20...tyle%20RV4.pdf

[kg]

It's not about data sheets, it's about best practices. As I'm sure you know, best practice does not abide by data sheets or any published literature, it's about objectively looking at the design and thinking "does this make sense". I'm a welding engineer, and pretty much everyday the text book gets left in the office while I play "best practices" making test welds.

well playing devil's advocate for a moment, if your best practice is that far off from the text book, shouldn't you wonder why?

Or to put it simply, my amp (or really test rig in this case), my rules. I've felt 500v, I don't ever want to again. And if that takes a more complicated solution that one already out in the wild, then I'll develop it with the help of people who know what they're doing.

sure, it's your amp your rules. but where does it end? shouldn't the entire amp be double insulated? with an isolation transformer input? with thermal cutouts? heck, let's run the whole thing at 40v b+, since it'll be safer.

[defaced]

well playing devil's advocate for a moment, if your best practice is that far off from the text book, shouldn't you wonder why?

I would say using a 500v part in a 500v environment is right at the limit of text book and where best practice kicks in. Kinda like using electrolytic caps right up to their voltage rating. Sure you can do it, but it's not a good idea because of the trade-offs involved. Designing to a limit is a good way to get bit.

sure, it's your amp your rules. but where does it end? shouldn't the entire amp be double insulated? with an isolation transformer input? with thermal cutouts? heck, let's run the whole thing at 40v b+, since it'll be safer.

For that matter, lets make everyone live in a bubble and feed them UV sterilized food and water...

It's not about being 100% safe, that's not possible. Every once in a while common sense kicks in and says "I don't care what "should" happen if something goes wrong because I "should" be safe, that looks like a really silly idea". This is one of those times, hence this thread and the question I asked in the very first post:

Am I being overly cautions or is this a really bad idea?

So, here's the 10 cent question: would you use this circuit on your amp?

[tedmich]

Defaced its about time you dropped the OTHER SHOE (tube wise) Go over to the extreme tubeophiles! Fully regulated voltages! Start with the Maida Regulator, and when you've blown up a few of those, try the Swenson version, then build one of Gary Pimm's designs:
The Swenson Regulator
I etched and built one of Gary's FQAF11N90 versions and it works great and is 900v capable

[guitarmike2107]

So, here's the 10 cent question: would you use this circuit on your amp?

yes.. but the components must be correctly spec'd to work in that design..... and the design could be "improved" a bit too with a few extra resistors.

This design works well for those with turret/eyelet board amps, if you are going to do a PCB then there is no reason not to add the more complex design

[Steve Conner]

heres the data sheet
http://media.digikey.com/pdf/Data%20...tyle%20RV4.pdf

So, those ones have an element voltage rating of 500V and a dielectric strength (I assume from element to case) of 900V. The latter is what's relevant when we're worrying about people getting zapped. In any case that can't happen if the panel is grounded. Most popular amps use less than 500V so problem solved!

KG, you can't really talk about best practice vs. textbook. Your BAGA looks nothing like a proper SLO clone.

[kg]

I would say using a 500v part in a 500v environment is right at the limit of text book and where best practice kicks in. Kinda like using electrolytic caps right up to their voltage rating. Sure you can do it, but it's not a good idea because of the trade-offs involved. Designing to a limit is a good way to get bit.

oh, i don't know about that.

no manufacturer worth their salt will rate a part at the 100% failure point. you can be sure that if a cap is rated to 500v, most will likely hold 1kv before popping.

will a cap last LONGER if run with LESS voltage across it? perhaps, yes, but there's no guarantee there either.

It's not about being 100% safe, that's not possible. Every once in a while common sense kicks in and says "I don't care what "should" happen if something goes wrong because I "should" be safe, that looks like a really silly idea".

well, to me, risk is the product of the probability of something happening, and the expected loss if it does happen.

in this case we reduce the probability of a fault occurring by using a properly specced part, with sufficient dielectric strength and voltage rating across the element.

we also reduce the expected loss by properly fusing the amp, and insuring that the chassis is grounded at all times, such that even IF the failure occurrs a nasty high voltage shock is not possible. if you wanted to be slick about it you could construct one of the simple three-LED style line circuit checkers (you know, the ones built into a plug that test for reversed neutral and proper grounding) right into the front panel of the amp. now you're arguably SAFER than the non-scaled but non line-tested amp.

by taking both steps, surely you agree that risk is minimized. whether or not it's sufficciently minimized is up to the designer. in this example, for you, clearly it is not, so you're forced to find some other method of (arguably more complicated) circuit design which reduces one (or both) of the risk factors.

So, here's the 10 cent question: would you use this circuit on your amp?

no, as i have no use for power scaling... unless it's to scale power UP.

[jrfrond]

There is NO F***IN' WAY I would hang a pot off of the B+ in tube gear!!! And then mount THAT to a CHASSIS???

UH-UH!!!

Part of the whole application of electronics is to control high voltage and/or high current using low-voltage control, and keep the operator as far away from the business end as possible.

Those of us who are old enough to remember, or simply savvy enough from restorations, will remember when B+ was sometimes located in external speaker cables when field-coil speakers were in use. I don't think you'd ever see such a risky maneuver ever again.

Yikes!!! :O

[defaced]

Defaced its about time you dropped the OTHER SHOE (tube wise) Go over to the extreme tubeophiles! Fully regulated voltages! Start with the Maida Regulator, and when you've blown up a few of those, try the Swenson version, then build one of Gary Pimm's designs:
The Swenson Regulator
I etched and built one of Gary's FQAF11N90 versions and it works great and is 900v capable

That's one hell of a page. I need to read more to understand it all, but that's some serious DIY engineering.

I searched for the FQAF11N90 part and I can't seem to find it - most places say it's out of production or obsolete. Fairchild lists the part as being in full production so I don't know what gives. Where did you purchase yours from or did you do a part substitution? Seems Fairchild makes plenty of HV parts, so I think getting a current production part is reasonable, which is something I'm looking at now.

[tedmich]

Hmm that mosfet does seem to be getting scarce... likely an STMicro STW12NK90Z
or IXYS IXFH12N90P from Mouser could be made to work

[J M Fahey]

Yes, Tedmich's suggestions should certainly work, MosFets are often generic.
I have been using them for a long time , and quite a few times the one I had been using dissappeared, under its old name, re-appearing under a new name, same specs (or *slightly* improved ones) , at a somewhat lower price.
Beauties of competition.
So, if specs match, go for it.
There's not rocket science at power scaling, we are talking very simple HV regulators here.