Plug and Play Power Supply

[exeric]

I'm gathering ideas for a plug and play power supply to use with all future amp builds. The idea presently is to have 3 regulated voltages, B+ = 0 to 800V, neg bias = 0 to -300V, and heater filament 0 to 12 V. I plan on using 3 dedicated transformers and 3 Maida regulators to implement it. The plan is to put the resistor that is used to determine each output voltage, (which is usually in the regulator) into each amplifier I build. This is the resistor in the output voltage divider that is closest to the 317.

My question is this: Is this feasable? That is, will the regulator oscillate if there is such a long wire going to this resistor in the amp? Also will it pick up noise that makes it impractical. There will probably be a Jones connector on the end of the umbilical that attaches to every amp. It seems to be the cheapest connector that can handle high voltages. And since it will go into every amp that will be the only modification to a standard amp build. The resistors for each voltage will be wired to that connector. The umbilical will be hard wired to the Power Supply on the other end. Presumably there will be a return wire also going back to the lower resistor in the voltage divider in the PS which doesn't help either. I'd really like to do this but don't want to get too excited if its just a pipe dream that can't be done. I know there's people here more knowledgable about the technicalities involved than I am.

I've decided I'm not a engineer so I will avoid deeply complex solutions to this possible problem. So keep that in mind.

Eric

[Wavebourn]

You may start from purchasing of some surplus laboratory PS that already has such options.

[exeric]

I didn't know they made such things. Do you have any examples?

[ethermion]

Fluke 407. HP, Heath, Krohn Hite, etc, all made reasonable ones too. B+, bias, and heater all out the front panel. B+ will max out anywhere from 350 to 550 volts. If you really think you need more than 500 volts, look for an electrophoresis power supply which will get you 2K+ volts. But anytime you go over 600V or so, you are trickier territory where regular wire and build techniques are not enough. Electrons over 600V are a little freaked out.

Careful shopping will snag one for under $100 delivered to your door. Ebay for tube bench power supply or the such. Very handy, and less likely to go poof. Millions of 'em made.

And pick up Valve Amplifiers by Morgan Jones, who explains in great depth power supplies in great depth, but is still an easy and enjoyable read for the non-engineer.

[exeric]

Actually I plan on getting one of those old power supplies that are used. I may not have been that clear about what I'm attempting to build. This will be an outboard power supply just like any pre-amp, amp, phono amp, that comes with an outboard PS. The trick is to build one of these outboard PS so it can be used with any amp, preamp etc that you build without having to adjust voltages.

To do that you have to the fixed resistors that determine the voltage INSIDE the amp you build. Its similar to what I used to do working in avionics for the airlines. Different airframes would have different options while using the same basic electronics. Many times those options were picked at the connecter and were hard wired for that airplane. Any box you plugged in would be automatically set for that airplane. Same principle here, though slightly different.

The whole idea for me (I'll let others decide if its a good idea-so far it doesn't seem to have caught on) is that the power supply in any amp is the most mundane and generic part of an build. Why go through all the trouble of build putting in nearly the same transformers, filters, regulators everytime you build an amp if you don't have to. It also frees up a lot of space. Of course you will have to probably put in sub-regulators within any amp for say screen voltage or whatever you dream up. However the basics will always be there with the right DC voltage for any amp. You can just take one amp out and plug another in and no worries about remember to set the correct voltages on the PS.

That's why I have to know if the outboard resistor on that long wire will make the regulator unstable. Anyone know?

Eric

[aardvarkash10]

Originally Posted by exeric

My question is this: Is this feasable? Eric

Wrong question. You should be asking "is this neccssary or practical?"

I can see how it might be a good idea in aircraft, I fail to see how it would be in this application. You would have to build a wildly over-specced supply to meet your design goal of wide capability. That means big upfront cost and complexity.

And if you ever wanted to sell any of the amps you are powering from it, they wouldn't have a PS, and therefore are, to all intents and purposes, a PoS.

JM2CW

[exeric]

Let me just go back to the essential reason for this post. I read a post by Sy somewhere where he said the upper resistor in the voltage divider in the Maida regulator had to be tied very closely to the LM317. He didn't explain that. I assumed the reason he said that because it would tend to oscillate otherwise, like grid damping resistors being tied close to the tube.

Really that's all I want to know at this point. I'm going to do this if that resistor isn't an issue because it makes sense for me. I don't sell amps of my own construction. I instead tend to canabalize parts for new construction. Its obvious it doesn't make sense to others. That's their problem. I'm not in this to follow the crowd or to be a sheep.

Can't anyone on this site get past their own prejudices and answer this technical question? You don't have to agree with my reasoning to give the answer to that question. Hey SY (if you're around) what's the story about what you said?

[chrish]

Or, you could build as you intend, place the set resistor right next to the chip, but allow for it to be easy to replace when required, maybe with a header style pin connector with the resistor in the plug, rather than at the end of a long lead.

[SY]

Originally Posted by exeric

Can't anyone on this site get past their own prejudices and answer this technical question? You don't have to agree with my reasoning to give the answer to that question. Hey SY (if you're around) what's the story about what you said?

Hier stehe ich.

In any case, the reason to tie that resistor close to the chip is for regulation. You want to minimize the drop (which is current dependent) between the output of the reg and that resistor, otherwise regulation is degraded, since that drop is effectively in series with the feedback.

The current through the lower resistor is quite small and not very load dependent, so you can use the bottom of that resistor for remote sensing- the extra wire (and its resistance) won't degrade regulation.

[exeric]

Thanks for the help. That's what I needed to know. All systems go, countdown commencing, engines on....

Eric