Power Supply Drop Resistor

[JJH0906]

Hey gang!

I am trying to get together my drop resistor network for my build but I'm having some trouble in getting things to work out. I put the diagram as shown in LTSpice and was working out the resistors and after a little trial and error, voila! I got the values I needed to drop the voltages down as necessary. However, when checking the wattage needed for each resistor, that's when things went wrong... very very wrong, like 20+ watts wrong!

Has anyone got any suggestions of how to calculate the drop resistors correctly? I am also not sure if my line diagram is the best way to represent what the power supply sees, I picked up on a similar diagram from one of Richard Kuenehl's books but his circuit wasn't the same as mine, so maybe I've got something mixed up somewhere.

[Matthews Guitars]

https://www.electronics-tutorials.ws/dc ... vider.html


There's no shortcut. Unless you've got a pile of high wattage resistor substitution boxes handy.

You must start by defining the voltage and current that will be drawn at each tap.

[JJH0906]

Yes, I did not put what I believe would be the current draw at each node, although I did write down the voltage. I did pre-calculate that and that's where I started putting the resistor values in, but like I said, there seem to be an excessive amount of wattage it some of the nodes. This is where the confusion started.

Actually the confusion started a few months ago when I started on this project...

[Matthews Guitars]

Just remember, you don't have to spec the resistor for more than the VOLTAGE DROP ACROSS THE RESISTOR over the resistance.

If you have series 10K resistors in a section of the power supply that's dropping 340 volts at the input to 320 volts at the next node and 300 volts at the following node, then you only have to figure the resistor size based on the 20 volt drop across each resistor. So if each tap needs 30 mA, then what's 30 mA at 20 volts? Easy. .6 watts. In this case 1 watt resistors are big enough.

[JJH0906]

Just remember, you don't have to spec the resistor for more than the VOLTAGE DROP ACROSS THE RESISTOR over the resistance.

If you have series 10K resistors in a section of the power supply that's dropping 340 volts at the input to 320 volts at the next node and 300 volts at the following node, then you only have to figure the resistor size based on the 20 volt drop across each resistor. So if each tap needs 30 mA, then what's 30 mA at 20 volts? Easy. .6 watts. In this case 1 watt resistors are big enough.

That is great info, thanks! I will sit down with this again to rework that.

It's crazy how we've got tube load lines, transformer calcs and all that but coming up with the right drop resistor network seems like such a mystery!

[Matthews Guitars]

When in doubt, just start with the power supply design of an amp that is generally similar to yours. Use its ladder values. And then without knowing Ohm's Law but with a selection of 1 watt resistors in standard values, you could just push resistor values one way or the other and see how the voltages change. Via this empirical method you could find your optimal resistor values to get the voltages you want.

Just remember that the choice of resistor values affects the current capacity of the power supply. You could just as easily drop 500 volts to ground using five large 1 ohm resistors as you could using five 1 megohm resistors. But with the 1 ohm resistors you'd have a high current supply and with the 1 megohm resistors you'd have almost no current capacity at all.

Using resistors that are in the general range of 5 to 15 K each gives you that balance of adequate current capacity.

[JJH0906]

This is all great information and it included me into what I did wrong. I naively was trying to drop the high voltage across the resistor, so for instance instead of dropping 340 to 320 volts, I was trying to drop 340 volts! Oops! I think with the advice that's been given here, I'm going to find my wattage ratings much more tolerable!

[Stevem]

You do not need to spend the bucks for a big 200 ma Choke with only two 6l6 outputs and that amount of preamp tubes.

I am also questioning why your using a AT7 for the reverb recovery stage and not a AX7?

[JJH0906]

You do not need to spend the bucks for a big 200 ma Choke with only two 6l6 outputs and that amount of preamp tubes.

I am also questioning why your using a AT7 for the reverb recovery stage and not a AX7?

Oops! Too late: 200 mA choke has already been procured. BUT! It knocks that AC ripple down to pretty much zilch without any filtering!

The choke circuit was based off an amp build video seen here: https://www.youtube.com/watch?v=eNBgo9Md8IU&t=1077s
The power circuit from that video looks as such:

KT88_BlueGlow_AmpBuildPowerSupply.JPG

The reverb circuit was based off of an amp "rebuild" at this video link: https://www.youtube.com/watch?v=tbGb74xLK9Q&t=764s
The associated revised reverb circuit from that video is:

Uncle Doug Supro TremoVerb Corrected 7 Nov 15 (1).pdf

That's where I got this information from and attempted to incorporate it into this build design.

[Matthews Guitars]

A very useful tool that I highly recommend any amp tech should have is a hand-held thermal camera. You can use it to quickly SEE where you have thermal problems due to undersized components and other thermal issues.

[pdf64]

… 200 mA choke has already been procured. BUT! It knocks that AC ripple down to pretty much zilch without any filtering!…

Class AB amps may draw much more HT current at high signal power output.
Hence at such high demand, a 200mA choke supplying the whole amp may saturate, lose its inductance, and the HT becomes very ripply just when we need it to be smooth. ie there may be intermodulation between the desired audio signal and the HT ripple, generating ‘ghost notes’.

[JJH0906]

Thanks. I'm reworking things and looking at the Valve Wizard site to get a better idea of what would work better.