I am building a tube amp (for guitar) from bits and pieces I have acquired over the years, including the main power (plate) transformer. The problem is that it is supplying more voltage than the circuit I am working on requires. I went with a tube rectifier and a choke-input filter scheme to reduce the voltage a bit, but I'm still about 100 volts too high. .
What would be the most elegant method for dropping 100V out of the supply (short of buying the right transformer)? A string of zener diodes, or a voltage divider?
One method would be to wire an unused filament winding in series with the primary. In one phase it'll reduce the secondary voltage, but maybe not by enough. You have about 400V and want about 300, is that right? A 12.6V winding will reduce the secondary voltage by only about 10%.
Another method is to use a centre-tapped full-wave rectifier if you aren't already--using a 5U4 or something like that.
A third method would be to make a buck switching supply with a 6146B switch. ;)
It's the B+ coming off of the rectifier, before the voltage divider that I am looking at. I have 500V, but it needs to be 400V; my thoughts were to keep the existing downstream components in the PS, and drop the volts before. I'm not sure of the load current, but the transformer specified in
How about a string or two of blue LEDs, mounted inside the cabinet. The glow will pulsate along with the sound, reminiscent of the voltage-regulator tubes of yesteryear.
Any kind of resistive solution is going to dissipate many watts and make the voltage regulation much poorer.
The right way is to switch to a choke-input filter (I'm assuming you have capacitor-input right now). That will drop the voltage a good
30%, and will actually improve the voltage regulation (plus make the revctifier tube much happier). If that's too much of a drop, put in a small input capacitor (just a few uF).
elegant if you're considering zeners that probably means engineering elegance - unless you consider neon lights under your car or in your computer box elegant . . . I'm guessing that's not you.
I think Phil Hobbs has your answer there. You wire a "bucking transformer" that will subtract the amount you need (percentage wise) from the primary. I don't know if he got the idea across . . .
The added transformer is wired to the power line on its primary so it produces AC voltage on its secondary - this voltage is wired in series with the plate transformer's primary but out of phase - subtracting voltage from the primary - the only thing you need is a transformer of the correct voltage and a SECONDARY that will handle the current (so it will be smaller physically than the plate transformer - just needs to carry the bucking current - and that will be less than the full plate xformer current since it will be working at reduced voltage . . .
Running at 120 VAC in and want to lower the output by 25% for example . . . you need a 30 VAC transformer (more or less). If your primary is pulling 2 amps your transformer needs to be 2 amps (probably 1.5 would be OK too)
That would be elegant from an efficiency point of view.
If elegant has to look impressive wire some incandescent lamps in series with the primary.
You didn't mention the filament supplies - if that's supplied by the plate transformer . . . wire the incandescent lights in series with the secondary HT supply only.
You could add a bucking transformer to the secondary HT supply only but that takes a relatively high voltage transformer which could be hard to find.
If it is a toroid transformer . . . wind the bucking winding directly on the outside of the other windings.
A small "home brew" un isolated switching supply would be easy if you have the knowledge. A switcher that lowers voltage can be nothing more than a comparator with some hysteresis so it turns on at a lower voltage and off again at a higher one. It turns the pass transistor, SCR, etc. (or hey for effects go for a big blue glowing mercury thyratron) and connects the output of the transformer to the filter cap until the voltage rises to the turn- off point, then turns off the pass device, until the voltage decays enough to turn back on. The concept is a piece of cake - the engineering isn't that bad - protect the low voltage parts from HV and choose a good pass element.
A simple phase controlled SCR after the rectifier before the filters would work. Gate has to be positive with respect to the cathode - that might take some thinking.\\
Pure inductance before the filter caps is another good way to lower the voltage reactively and efficiently - put a transformer primary in series with the secondary.
Remember - some of these ideas require current to be flowing before they work - and they won't start lowering voltage until the filaments are hot and drawing current on the cathodes. - For maximum toob life you want the filaments hot before you goose them with plate voltage - a time delay relay would be a good idea.
Have fun with it.
----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
formatting link
The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Well.......... You need to drop 100V @ up to 90mA. That means 9W of dissipation. Although the B+ input will drop a bit when loaded @ 90mA, so not quite 9W in reality.
I'd be tempted to use a voltage regultor here actually for a 'solid' ripple free
400V but that may upset any deferrence to 'toobiness'.
I'd use a Mosfet as a series pass element. It would work a treat and just need a 'medium size' heatsink. Getting rid of the heat will be an issue whichever way you do it.
Is that measured under load? With a choke-input filter you'll see a drop from 1.4*secondary at no load (the choke needs current flow to do its work!) to 0.9*secondary at some minimum load (typically 1000/choke inductance to give minimum load in mA) and then at higher loads you get damn good regulation.
Under full load you'll also be seeing some voltage drop from the rectifier tube and choke resistance.
Bucking transformers, as others suggested, in either primary or secondary.
Or use this as an opportunity to learn how a different B+ picks a different load line for your tube and changes the output impedance etc.
Thanks for the extra explanation, default, I needed that (been awhile since I played with bucking trannys). This also solved the other problem I discovered today, that the filament voltages were too high - no doubt due to both the typical line at the time the power transformer was built and my
125VAC house line.
I used a 16V tranny I had in the parts drawer to drop the line to 108VAC; this put the filaments at 6.35V (and the rectifier filament at 5.2V, just four percent higher than it calls for, and I can live with that).
Right now the B+ is sitting at 430VDC, which, considering the rest of the circuit isn't hooked up yet, is in the ballpark enough for me to build the rest of the amp, then bring it up on the variac and refine the power supply components at that point.
Thanks again for all of the help; it is *really* appreciated.
Why not add a bucking transformer winding in series with the ac line. that way you won't affect the regulation (or lack of it) and possibly upset any deference to toobiness. a 24 volt three amp filament transformer should do nicely. regards, tom
I see no load in this circuit. A choke input to filter MUST be under load to be effective. If there is insufficient current, the reactance of the choke does nothing and the circuit acts like a capacitor input to filter; i.e., it charges to the peak value as though the choke is not there. Be sure to measure it under load. If there is insufficient load you must use a bleeder resistor to insure proper loading under all conditions. At 500 volts it sounds like it may be charging to near the peak without choke averaging. What is the voltage with the choke shorted? That will give you a clue as to whether you ar getting averaging or not.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.