Looking for Supro Thunderbolt 6420 Build info

[tubeswell]

I've used a shared 10W 270R cathode resistor in a few t-bolt style builds. Bypassing with 25uF yields more grunt.

For the screen grids, allow for 5-6mA idle screen current per tube for 5881s or 6L6s. 1W to 3W 470R resistors are ample for this, and should ensure reliable functioning, even when the screens pull maximum current under heavy signal conditions. All they are required to do is eat up a bit of current when the screens are running hot. (The '5W' screen power rating on 6L6GC tube data sheets is the maximum Pg2 rating. So if you use 5W resistors, there's a possibility the screens will burn up before the Rg2s will, which could cause other problems)

If you go smaller (like 1/2W), the amp will still run fine, but if the screens get too hot, the resistors will burn and go open more easily. Some guys prefer this so that the Rg2s act like a fuse when the screens run onto trouble. Therefore, if you're going to do that, the Rg2s should be the flame proof (fusible) type . However, replacing blown Rg2s is a real PITA if you have do do it all the time, so its up to you.

Another alternative (when idling the plates at full Pmax) is to run the screens about 50V below the plates, through a largish supply dropping resistor (5k-10k) like early Fender tweed amps. This will ensure the grid curves are squashed down sufficiently far enough that the load line runs through (or above) the knee of the Vg=0V curve, thereby preventing large spikes in Ig2 when the amp is being driven hard.

[eschertron]

Awesome, thanks!

That was my first attempt at reply from a mobile device. I hate the limited screen real estate, and the clunky interface. Makes my response way more terse than it needs to be.
I assume the 1k resistor is at the PSU screen node, and not individual screen stoppers. The schem doesn't want to open for me.

[Helmholtz]

You calculated the power the resistor has to pass. You want the power the resistor keeps to itself and converts to heat. This is given by I²R.

[mikepukmel]

You calculated the power the resistor has to pass. You want the power the resistor keeps to itself and converts to heat. This is given by I²R.

Ok thanks Helmholtz. So its two pieces: the part of charge that moves downstream, and the part inside this 1k dropping resistor. Since I don't have the amp built, I can't measure the voltage drop, but maybe can guess using the Weatherford numbers.

EDIT: Oumpf, sorry, I²R, can guess I, and have R, so don't need the drop.

[mikepukmel]

That was my first attempt at reply from a mobile device. I hate the limited screen real estate, and the clunky interface. Makes my response way more terse than it needs to be.
I assume the 1k resistor is at the PSU screen node, and not individual screen stoppers. The schem doesn't want to open for me.

Oh man I gave up my cell phone a little over 1 1/2 years ago. Miss it some, when I need to get in contact with someone, but don't want to wait till I get home, but don't miss that teeny screen!

[mikepukmel]

I've used a shared 10W 270R cathode resistor in a few t-bolt style builds. Bypassing with 25uF yields more grunt.

For the screen grids, allow for 5-6mA idle screen current per tube for 5881s or 6L6s. 1W to 3W 470R resistors are ample for this, and should ensure reliable functioning, even when the screens pull maximum current under heavy signal conditions. All they are required to do is eat up a bit of current when the screens are running hot. (The '5W' screen power rating on 6L6GC tube data sheets is the maximum Pg2 rating. So if you use 5W resistors, there's a possibility the screens will burn up before the Rg2s will, which could cause other problems)

If you go smaller (like 1/2W), the amp will still run fine, but if the screens get too hot, the resistors will burn and go open more easily. Some guys prefer this so that the Rg2s act like a fuse when the screens run onto trouble. Therefore, if you're going to do that, the Rg2s should be the flame proof (fusible) type . However, replacing blown Rg2s is a real PITA if you have do do it all the time, so its up to you.

Another alternative (when idling the plates at full Pmax) is to run the screens about 50V below the plates, through a largish supply dropping resistor (5k-10k) like early Fender tweed amps. This will ensure the grid curves are squashed down sufficiently far enough that the load line runs through (or above) the knee of the Vg=0V curve, thereby preventing large spikes in Ig2 when the amp is being driven hard.

Awesome, thanks reading, re-reading. So, pushing the screen grids down, this will produce lower power, but also, will this cause more of that "old amp" tube compression?

[mikepukmel]

This would be fine. If you want, you could build it with both SS rectification (2-phase rectifier) and tube rectification, so you can swap between them and get the different characteristics.

Fantastic, yep I wanted to build this a SS rect amp for a couple of reasons. Curious about the sound and didn't want to wire up another tube rectifier, as I have 2 tube rect amps already. And maybe a bit of blasphemy, but one fewer tube in this amp!

[tubeswell]

Ok thanks Helmholtz. So its two pieces: the part of charge that moves downstream, and the part inside this 1k dropping resistor. Since I don't have the amp built, I can't measure the voltage drop, but maybe can guess using the Weatherford numbers.

A good rule of thumb would be 1mA per 12AX7 triode and 4-5mA per 6L6 screen. See the diagram at the bottom of Merlin’s page here (albeit that it uses EL34s)

http://www.valvewizard.co.uk/smoothing.html

[mikepukmel]

Arlyn from Heyboer got back to me, I had sent an email asking about his HTS-4664 power transformer, that is similar to this one, but doesn't have the 5v heater tap which this ckt doesn't need. Its comparable price to the 2 others, the Hammond and the Classictone. The Classictone is a lot higher voltage on the HV winding, 350.

[tubeswell]

Yes

[Helmholtz]

I don't think using idle screen currents is safe enough - especially in a guitar amp. The GE datasheet specifies a maximum signal screen current (DC average) for two 6L6-GC of 22mA in an amp with Vp=450, Vs=400V.

To find an upper limit (in a well designed amp) for the screen current per tube I suggest to use max. screen dissipation and divide by screen voltage at full output (maybe 10 to 15% lower than at idle depending on PT regulation and screen resistors).

[mikepukmel]

I don't think using idle screen currents is safe enough - especially in a guitar amp. The GE datasheet specifies a maximum signal screen current (DC average) for two 6L6-GC of 22mA in an amp with Vp=450, Vs=400V.

To find an upper limit (in a well designed amp) for the screen current per tube I suggest to use max. screen dissipation and divide by screen voltage at full output (maybe 10 to 15% lower than at idle depending on PT regulation and screen resistors).

Thanks Helmholtz. For the values "at full output, or 10% below" is there a way to determine/estimate the numbers before the amp is built, i.e. amp not built yet so can't measure directly?

[mikepukmel]

Got some parts 'roughed' in, not soldered and trimmed yet. Installed a few more turrets to make the mounting distance small enough for these new teeny resistors. (dang, dug through both boxes of parts, ordered 2 x 270k resistors and no 3 watt 27k)

A lot more sparse than I thought, based on my hand layout drawing, even though some components are a bit cramped over by V1 preamp tube. OCD is telling me to tear everything off, order a blank piece of fiberglass board and straighten things out a bit.

[mikepukmel]

Question: plenty of room to move the two power supply caps feeding the preamp tube and phase inverter tube, next to where they are connected to the plate resistors. Is it better to have those parts closer to the plate resistor, with a longer wire between the previous power supply dropping resistor/cap, or have the cap/resistor closer to the other power supply parts? Is one better than the other?

[Helmholtz]

Thanks Helmholtz. For the values "at full output, or 10% below" is there a way to determine/estimate the numbers before the amp is built, i.e. amp not built yet so can't measure directly?

This might be a misinterpretation. My proposal is to use screen voltage at idle -15% to account for sag and some voltage drop across screen resistors. Dividing screen dissipation by a lowered voltage gives a higher max. screen current and thus results in a higher power rating of the dropper resistor.
The principle behind assumes that the amp was designed to use up full screen dissipation at full power, taking into account reduced voltages at full power.

Example: max. screen dissipation = 5W, screen node dropper = 1k, assumed screen voltage at full power = 425VDC. This gives 2x5W/425V = 23.5mA max. screen current for 2 tubes. Addind a few mA for the preamp tubes, we can assume that the dropper will not see more than 30mA, resulting in 0.9W.
Now if the amp is going to be subjected to longer periods of power stage clipping, I would use a 2W resistor.