Simulating 6G2 bias circuit -- am I close?

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
Heh, I'm a total infant at circuit simulation. Here's my 6G2's simple bias circuit (a variant of Doug Hoffman's time-honored design) with some current measured voltages in red... (Edit: diagram corrected to show the current 15K tail resistor).

1671507269319.png


And here's my attempt to model it on falstad.com, an online simulator:

falstad 1.png


(I didn't draw in the failsafe jumper, a recent addition, but if I understand, it doesn't change function when the wiper is working). What did I miss, or do wrong? I note in blue that I set the PT max voltage at 390=275*1.414, and you can see the V(rms) before the range resistor and before the diode, and the (± steady) Vmax before the pot. So these are 'kinda close' to the measured values -- but I can readily picture I left out all sorts of important elements or settings.
 
Last edited:

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
? Resistor value to ground in your schematic...15K vs 27K in the layout.
Well spotted, sir. My mistake; grabbed the wrong version of that diagram. That's the original, copied from my Princeton Reverb. I edited the OP to the current version to decrease confusion. Thanks for the heads up.

Looks about right. Try it and see
Thank you, tubeswell; do you mean the model is adequate and correct for the actual circuit? If I didn't make it clear, first I 'tried it' in the build, and it works well; today I modeled it and wanted to check if I got the features and values right, especially since the model voltages are only 'kinda close' to the measured.
 

Ten Over

Tele-Afflicted
Joined
May 13, 2015
Posts
1,517
Location
Central California
(I didn't draw in the failsafe jumper, a recent addition, but if I understand, it doesn't change function when the wiper is working). What did I miss, or do wrong?
The jumper changes it to a different kind of circuit. With the jumper, the total load changes when the wiper moves. Without the jumper, the total load is the same when the wiper moves.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
The jumper changes it to a different kind of circuit. With the jumper, the total load changes when the wiper moves. Without the jumper, the total load is the same when the wiper moves.
Thanks. You've shared this concept before, but I wasn't quick to see the difference. It's starting to sink in; I took a stab at correcting the simulation. Closer?

1671511842194.png
 
Last edited:

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,634
Location
Crawfordville, FL
@King Fan, I believe you want to simulate that using the unloaded secondary voltage, which would be closer to 300Vdc depending on your mains voltage. The positive B+ supply rectifier is only working with the positive half cycles from each half of the secondary. Since there's no current 'flow' on the negative half cycle, there's no voltage drop across the source resistance. I could be wrong.
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,634
Location
Crawfordville, FL
I started to say that you may want to simulate the source resistance anyway, but I realized that it's just a tiny drop in the bucket compared to that 100k resistor, so you can just ignore it.
 

peteb

Poster Extraordinaire
Joined
Apr 25, 2003
Posts
5,573
Location
Cascadia
interesting simulation!

(I didn't draw in the failsafe jumper, a recent addition, but if I understand, it doesn't change function when the wiper is working). What did I miss, or do wrong? I note in blue that I set the PT max voltage at 390=275*1.414, and you can see the V(rms) before the range resistor and before the diode, and the (± steady) Vmax before the pot. So these are 'kinda close' to the measured values -- but I can readily picture I left out all sorts of important elements or settings.

which values don’t look right to you?

what is the green AC versus the yellow AC?


is there a measurement in between the dropping resistor and the diode?
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
@King Fan, I believe you want to simulate that using the unloaded secondary voltage, which would be closer to 300Vdc depending on your mains voltage. The positive B+ supply rectifier is only working with the positive half cycles from each half of the secondary. Since there's no current 'flow' on the negative half cycle, there's no voltage drop across the source resistance. I could be wrong.

I started to say that you may want to simulate the source resistance anyway, but I realized that it's just a tiny drop in the bucket compared to that 100k resistor, so you can just ignore it.

Thanks, Phryg. Source voltage was one of a number of things I guessed at -- I chose the 1.414 X (nominal) 275 = 390 because even at that level the simulation was a bit low -- 275Vrms where measured voltage was 282-287. As I say, a chlld lost in the woods. :) I like your second point, but if I could model source voltage more realistically, I'm all ears.

Let me try to clean up the original display. I'll leave out the yellow current traces, and the jumper, too -- I measured those voltages before I added it.

1671639888758.png
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,634
Location
Crawfordville, FL
@King Fan what PT are you wanting to simulate? I assumed a Hammond that has the 550V secondary or taps, either the one specifically for the Harvard (290HAX) or the PR type that has taps (290AX). Those are 296.4 and 297.3 Vrms at 120 Vrms input respectively. The latter is going to be about 309.7 Vrms at 125 Vrms in, or 437.97 Vpk.

Note that it takes several seconds for the cap to charge up.

 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
Thanks, @Phrygian77 . The PT is actually the good ol' CT 40-18027, 275-0-275 @ 100mA. My approach to setting output Vpk has been total SOP = seat of pants. I just guess, then adjust it until I see something like the 287V measured before the range resistor. Here's how clueless I am -- I'm just assuming my measurements are Vrms. And FWIW falstad.com doesn't offer a 1N4007, so I'm just using the 'default-spice' diode.

The only place I can claim even one clue is I do wait for the cap to charge up and the voltages to reach steady state. I try to adjust the sim speed to be fast but readable.

At this point I'm getting tired of scrolling up to see the measured voltages. Time to copy and paste (but I'm lazy; I still didn't delete the jumper to match my measured state.)

1671648952347.png
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
Oh, and you remind me I should be able to link to my actual sim so folks can interrogate it... let's try this. Yep, click run and have at it -- my saved version shouldn't break if folks tweak this one.

 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,634
Location
Crawfordville, FL
@King Fan measure the PT unloaded. I suspect it's similar to the Hammond. I think you're going to get 295 to 300 Vrms, so you you're most likely short 20-25 volts on your sim.

One thing I noticed that I hadn't really thought about before, I guess because I haven't built a 6G2, is that there's something fishy about Fender's -35V bias spec on the schematic. If the B+ is actually 315V, -35V is ice cold, approaching cutoff cold. Plate voltage in the '62 that I had on my bench was 385V.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
9,481
Location
Salt Lake City
Y=
@King Fan measure the PT unloaded. I suspect it's similar to the Hammond. I think you're going to get 295 to 300 Vrms, so you you're most likely short 20-25 volts on your sim.

One thing I noticed that I hadn't really thought about before, I guess because I haven't built a 6G2, is that there's something fishy about Fender's -35V bias spec on the schematic. If the B+ is actually 315V, -35V is ice cold, approaching cutoff cold. Plate voltage in the '62 that I had on my bench was 385V.

Thanks, yeah, I did these measurements as a sideline while setting up the bias with the new 15K tail resistor to increase the range. I'll have to get back in there with more attention to the voltages.

Luckily for me, I never pay much attention to Fender bias voltages. I just calculate %MPD and aim for a range that gets me at least 50% to 70%. The bias measurements I can do from outside the amp, so I always double check when I get it buttoned back up. But now I wonder if I measured something wrong on that bias board. Hey, I know, I'll check it next year! :)
 
Top