6G3 Capacitor Question

[dwinstonwood]

66Strat, this is all starting to go way over my head. I'll need to back up and review what I do understand.

I've only built cathode biased amps, never a fixed bias type. So, I really don't know what it means to "estimate a lower tube current." I understand that I divide plate voltage (285V) by the plate current (.035A) to get dissipation in watts, but I don't know how to manipulate plate current in this type of amp.

So, are you saying that a plate voltage of 285V is too high, or is it the target I should aim for? And, should this be calculated with a signal or at idle? In other words, are you saying that the tremolo in the 6G3 will act weird if the PT supplies more than 285V to the plates unless I change the plate current?

Sorry, if I'm missing something obvious. I feel like I have to learn a whole new set of things with every new project.

[dwinstonwood]

Using an online bias calculator I got 29.5ma for 70% of 6V6 dissipation. What I've never done and don't know how to do is achieve 29.5ma of plate current. I have no idea which components to change to get to this figure.

Or, do I need to lower the plate voltage to do this. 240V plate voltage and 35ma would hit the 70% target.

[66Strat]

66Strat, this is all starting to go way over my head. I'll need to back up and review what I do understand.

I've only built cathode biased amps, never a fixed bias type. So, I really don't know what it means to "estimate a lower tube current." I understand that I divide plate voltage (285V) by the plate current (.035A) to get dissipation in watts, but I don't know how to manipulate plate current in this type of amp.

So, are you saying that a plate voltage of 285V is too high, or is it the target I should aim for? And, should this be calculated with a signal or at idle? In other words, are you saying that the tremolo in the 6G3 will act weird if the PT supplies more than 285V to the plates unless I change the plate current?

Sorry, if I'm missing something obvious. I feel like I have to learn a whole new set of things with every new project.

To calculate plate dissipation, you multiply plate current times plate voltage. (285 x .035 = 9.975 watts) In a fixed bias amplifier, to adjust plate current, you adjust the negative bias supply voltage by either substituting resistors in the voltage divider of the bias supply circuit or implement a bias voltage pot.

IF one chose to operate at a plate voltage of 365 volts (375 - 10 volts output transformer IR loss), 35 ma would result in 12.775 watts plate dissipation (91.25% of design maximum of 14 watts). This is on the hot side of things for class AB1 operation and too hot for the tremolo to work properly IMO. Seventy percent of the Design Maximum plate dissipation of 14 watts or 9.8 watts would be a better target. The 9.8 watts plate dissipation divided by the plate voltage of 365 equals 26.85 milliamps of plate current, plus 2 milliamps for screen current equals 28.85 ma current for each 6V6 tube. To measure the current, measure the voltage across the 1 ohm cathode resistor of each 6V6 tube with the volt ohm meter set to millivolts. Each millivolt of cathode voltage equates to 1 milliamp of tube current. For 70% plate dissipation at 365 plate volts, the cathode voltage reading would be 28.85 millivolts across the 1 ohm resistor. To adjust the bias current adjust the negative bias voltage until you reach the 28.85 millivolts across the cathode resistor.

Keep in mind that the plate voltage will move some as you manipulate the bias voltage. This is where the rectifier graphs that I previously posted come into play as a starting point to estimate what the expected DC voltage will be at an estimated current. Use the charts as a starting point to establish the operating conditions and then fine tune the circuit by adjusting the bias voltage.

The starting points would be about 360 DC supply volts at 65 milliamps with the 270EX transformer and a 5AR4 rectifier or about 400 DC supply volts at 60 milliamps with the 272DX transformer and a 5U4 rectifier. Back out 10 volts or so to account for the IR loss in the output transformer and then adjust the negative bias voltage until the cathode current reads 30 ma for the 270EX 5AR4 combination or 27 ma for the 272DX 5U4GB combination.

[bmccowan]

this is all starting to go way over my head. I'll need to back up and review what I do understand

dwinstonwood, in my opinion you can make this as simple or as complicated as you want. If you are wanting to build a 6G3 (and I think you still are) - great choice. There are confirmed schematics and scores of folks without the textbook type knowledge have built that amp and many have added adjustable bias and have kept the bias wiggle type tremolo. I always get lost if I jump into details before I understand basics. You say that you are missing a basic understanding of how fixed bias amps work. I suggest that you read Rob Robinette's pieces, and/or others' on that subject. And then you can dive as deep as you want into charts, calculations, modelling, etc. I am not taking away from anything that 66 is stating - he knows a lot of technical details that I do not, but I sympathize with your

[dwinstonwood]

Thanks bmccowan. I have confidence that I can learn the basics of this stuff, but like you said, it's confusing diving into the middle of it. I actually did read Robinette's info last night and it cleared things up a good bit.

Thanks 66Strat. I was tired and frustrated last night and typed a bunch of errors. I meant to say 365V and to multiply ma by plate v to get dissipation. I was also mistakenly looking at 6V6's (12w max), not 6V6GT's (14w max).

So, I now understand the basic workings of biasing the amp. I can grasp how making the grid voltage more or less negative in relation to the cathode will adjust the idle current.

What I am starting to glimpse here is that finding a currently available PT is going to be a tightrope act involving the best rectifier choice and the correct negative bias voltage on the grid. So, this might mean disregarding the numbers on the Fender schematic to some extent in order to hit the 70%, 10 watt dissipation target.

Having said all that, I'm fine with running the 6V6GT's at 60%, or even less - that is, creating more leeway as opposed to trying to hit 70% on the nose. A PT/rectifier combo resulting in above or below 375 plate voltage if fine with me, if it works.

[66Strat]

Thanks bmccowan. I have confidence that I can learn the basics of this stuff, but like you said, it's confusing diving into the middle of it. I actually did read Robinette's info last night and it cleared things up a good bit.

Thanks 66Strat. I was tired and frustrated last night and typed a bunch of errors. I meant to say 365V and to multiply ma by plate v to get dissipation. I was also mistakenly looking at 6V6's (12w max), not 6V6GT's (14w max).

So, I now understand the basic workings of biasing the amp. I can grasp how making the grid voltage more or less negative in relation to the cathode will adjust the idle current.

What I am starting to glimpse here is that finding a currently available PT is going to be a tightrope act involving the best rectifier choice and the correct negative bias voltage on the grid. So, this might mean disregarding the numbers on the Fender schematic to some extent in order to hit the 70%, 10 watt dissipation target.

Having said all that, I'm fine with running the 6V6GT's at 60%, or even less - that is, creating more leeway as opposed to trying to hit 70% on the nose. A PT/rectifier combo resulting in above or below 375 plate voltage if fine with me, if it works.

Bingo!

There are some numbers on the Fender schematic that can be used to back in to an operating point. Others appear to be artifacts left over from a different transformer that Fender may have experimented with prior to production. The voltage under load numbers are very similar to 6G9 power transformer voltages.

This reminds me of a motorcycle safety course that I took some years back. The instructor taught us to avoid target fixation, and to look through the curve at where we wanted to go. On a bike, you go where you look. If you look at the ditch, that's where you wind up. Don't fixate on achieving the 375 V DC number. Any DC plate voltage between 350 and 400 will work well in the circuit. You just have to adjust the bias according to what you have.

[bmccowan]

This reminds me of a motorcycle safety course that I took some years back. The instructor taught us to avoid target fixation, and to look through the curve at where we wanted to go. On a bike, you go where you look. If you look at the ditch, that's where you wind up. Don't fixate on achieving the 375 V DC number. Any DC plate voltage between 350 and 400 will work well in the circuit. You just have to adjust the bias according to what you have.

Spot on advice - I too ride motorcycles and got the same advice. When I took up mountain biking - it was the same. A friend of mine races and teaches pro rally off road driving - same advice. Look where you want to go.
Have fun building the amp!

[shooter]

avoid target fixation

works with rally cars also 
I call it "the line", it's the brain doing complex math without numbers, an hopefully without trees n ditches and...... 

[dwinstonwood]

I got the exact same advice, too! But for snow skiing. Same principle. When you stop looking at the tips of the skis and start looking at the path ahead you stop wiping out.

66Strat, thanks for the starting point calculations. The 270EX/5AR4 combo looks good to me.

The one blindspot for me is how you arrived at 65ma? Why not 60ma and 370V, for example. You mentioned "28.85 ma current for each 6V6 tube." What am I missing here?

[66Strat]

This reminds me of a motorcycle safety course that I took some years back. The instructor taught us to avoid target fixation, and to look through the curve at where we wanted to go. On a bike, you go where you look. If you look at the ditch, that's where you wind up. Don't fixate on achieving the 375 V DC number. Any DC plate voltage between 350 and 400 will work well in the circuit. You just have to adjust the bias according to what you have.

Spot on advice - I too ride motorcycles and got the same advice. When I took up mountain biking - it was the same. A friend of mine races and teaches pro rally off road driving - same advice. Look where you want to go.
Have fun building the amp!

One of my bucket list items is to go to the salt flats during speed week. I would love to ride my bike as fast as I can go on the flats. When I was a sophomore in high school a couple of friends and I rode our bikes down to Myrtle Beach. My parents went to Disney World on vacation and I stayed home. They asked what my plans were. I said that that I might go camping, and I did. My friends and I camped on the beach amongst some dunes next to a large family campground. We used the showers and toilet facilities at the campground. There was a beach club that we went to at night. The Friends of Distinction and Junior Walker & the All Stars were appearing at the time. Both of these songs were on the radio at the time.
https://www.youtube.com/watch?v=-UHsjvPOZ3c
https://www.youtube.com/watch?v=ub72eylahJg

That was one of the best vacations I ever had.

I meant to ask earlier, what transformers/voltages did you use/get in your 6G3?

[66Strat]

I got the exact same advice, too! But for snow skiing. Same principle. When you stop looking at the tips of the skis and start looking at the path ahead you stop wiping out.

66Strat, thanks for the starting point calculations. The 270EX/5AR4 combo looks good to me.

The one blindspot for me is how you arrived at 65ma? Why not 60ma and 370V, for example. You mentioned "28.85 ma current for each 6V6 tube." What am I missing here?

If you look at the 6G3 schematic, it shows a plate voltage of 365 volts. Divide the target 9.8 watts plate dissipation by the plate voltage and you arrive at a plate current of 26.85 ma, adding 2 ma for screen current gets you to 28.85 ma. of tube current for each 6V6. Current for two 6V6 tubes is 57.7 ma add in 6 ma of 12AX7 tube current (1 ma for each triode) gets you to 63.7 ma. I just round up to 65 ma. The intersection of the 65 ma line with the blue  AC line gets you to the DC supply voltage of 360 volts. Subtracting 10 volts to account for the output transformer IR losses gets you to an anticipated plate voltage of 350 V DC.  Subtracting 10 ma (4 ma for the screens + 6 ma for the AX7s) from 65 ma gets you to 55 ma of plate current for two 6V6 tubes (27.5 ma each tube). Plate dissipation at this operating point would come in at 9.625 watts or 68.75% of the 14 watt maximum plate dissipation limit.

[dwinstonwood]

Perfect explanation! I forgot about the 12AX7 current. And the 68.75% range sounds great.

Well, it looks like you solved the difficult part of updating this amp 66Strat.

I could save some money by buying a chassis and punching it myself. I could use part of those savings to buy this and then save more money down the road:
https://store.triodestore.com/punchkit.html
That way, I can use an upright PT without the large hole in the custom 6G3 chassis.

[shooter]

adding the 1 ohm cathode resistors makes all these #'s easier, lots easier!

I would love to ride my bike as fast as I can go on the flats

Once I dialed in driving my Evo, took it to TN, found a hair-pin twisty straight up, 180 turn, twisting turning down. hit ~115 up, running gears down fast for the apex and there was a TN State Trooper just parked!  he just pointed down with his finger, I nodded thanks, as I past, he said, "watch out for the 3rd turn".  Sure enough, I read it wrong, but his warning saved me from a 200' short-cut 

commissioned an artist friend to draw this up for me 

[66Strat]

Perfect explanation! I forgot about the 12AX7 current. And the 68.75% range sounds great.

Well, it looks like you solved the difficult part of updating this amp 66Strat.

I could save some money by buying a chassis and punching it myself. I could use part of those savings to buy this and then save more money down the road:
https://store.triodestore.com/punchkit.html
That way, I can use an upright PT without the large hole in the custom 6G3 chassis.

Keep in mind that I am not advocating this transformer/rectifier combination and that the numbers represent best guesstimates to get in the ball park as a starting point for this combination.

You might check this guy out. He will make a chassis to your specs, including all punch-outs and holes. It might be worth your while to sketch out the chassis dimensions and layout for tubes and control holes on his order form and email it to him for a quote. He also does faceplates. I haven't used him, but his products look promising.
http://www.juicyamps.com/chassis.html

adding the 1 ohm cathode resistors makes all these #'s easier, lots easier!

I would love to ride my bike as fast as I can go on the flats

Once I dialed in driving my Evo, took it to TN, found a hair-pin twisty straight up, 180 turn, twisting turning down. hit ~115 up, running gears down fast for the apex and there was a TN State Trooper just parked!  he just pointed down with his finger, I nodded thanks, as I past, he said, "watch out for the 3rd turn".  Sure enough, I read it wrong, but his warning saved me from a 200' short-cut 

commissioned an artist friend to draw this up for me 

Tail of the Dragon?

ISP passed me while i was doing 110 on I-65. He shook his head and gave me the same slow down motion.

[bmccowan]

I meant to ask earlier, what transformers/voltages did you use/get in your 6G3?

It was a ClassicTone PT - and I remember being confused by the spec sheets  I do not recall which model, but I'll look when I get home, as I'm away for the weekend. I think it was one of the ones with 2 secondary taps and I used the lower voltage red/white taps. But can't trust the memory. I was thinking I built it in 2017, but I found the chassis order and it was 2013!

Now old and foolish, I ride a Moto Guzzi and am pretty happy staying under 80mph these days. She'll do more, but the top end starts sounding like a washing machine full of nuts and bolts.

[dwinstonwood]

bmccowan, yes, please post the specs for your PT when you get a chance.

I'm looking into Edcor again. I got great service from them and a good quality product. Here's two that might work.

XPWR224: 540V(270-0-270)@125mA, 6.3V(3.15-0-3.15)@3A, & 5V@2A
https://www.edcorusa.com/xpwr224

and,

XPWR163: 550V(275-0-275)@150mA, 6.3V(3.15-0-3.15)@5A & 5V@3A
https://www.edcorusa.com/xpwr163

I will attempt to plot 66Strat's graphs tonight with these two.

[shooter]

I been using Edcor >10yrs, nice stuff

while you're learn'n curves, keep you rec tube 5V Amps in mind, you listed a 2 and 3A

[dwinstonwood]

Thanks shooter. That's another spec I don't know how to figure out: not sure how to use those amperage numbers or what they mean exactly.

[dwinstonwood]

OK, another PT option! Sorry for posting so many.

Any reason why the Hammond Harvard 290HAX replacement wouldn't work? It's 275-0-275 @ 100ma, has a 50v bias tap, and 6.3v center tap:

https://www.hammfg.com/files/parts/pdf/290HAX.pdf

Thanks

[66Strat]

OK, another PT option! Sorry for posting so many.

Any reason why the Hammond Harvard 290HAX replacement wouldn't work? It's 275-0-275 @ 100ma, has a 50v bias tap, and 6.3v center tap:

https://www.hammfg.com/files/parts/pdf/290HAX.pdf

Thanks

It could work. The heater and rectifier currents are OK. The secondary current rating is marginal. As a result, the B+ supply voltage will sag more under load. But there is nothing inherently wrong with that, or anything there to cause the Earth to spin backwards, or cause the magnetic poles to reverse. It's all a mater of personal taste. I would encourage you to plot the dynamic load lines of any transformer that you consider. It is a helpful learning experience.

[dwinstonwood]

Thanks 66Strat. Will do!

Comparing the Harvard 5F10 to the Vibrolux 5E11 (and, 5F11), they are remarkably similar circuits in component values and stated voltages. I'll bet Fender used the same PT for these amps, and probably others, too, maybe even the 6G3. All three of these are 6V6GT amps with plate voltages of 365V or less. The 5F10 and 5F11 used 5Y3GT's and a different PI than the 6G3.
Years ago I really wanted a Harvard because I read that Cropper used one in the Stax studio. From what I can see, the Tweed Vibrolux is a Harvard with tremolo. Now, that's looking interesting...

[66Strat]

As long as you're investigating other tremolo amps, check out the 5E9-A Tweed Tremolux. This amp mates the preamp of a 5D3 5E3 Deluxe with the phase inverter and power amp topology of a 5B3 Deluxe. The tremolo wiggles the bias of the phase inverter.
https://el34world.com/charts/Schematics/files/Fender/Fender_tremolux_5e9a.pdf
https://www.youtube.com/watch?v=-Rmqaqe0y7k
https://www.youtube.com/watch?v=1cxoi8P2XSI


Edited to change 5D3 to 5E3

[shooter]

That's another spec I don't know how to figure out

66Strat was using a 5U4 example, that tube needs a 5V winding at 3A, a 5Y3 requires a 5volt winding at 2A.  that data is on the tube datasheets
SS rectifiers don't need an extra winding

[dwinstonwood]

Cool. But, I'm going to try and stay the course with the 6G3. In that first video, near the end, sounded like X's Billy Zoom playing Hungry Wolf. Maybe I need a Sparkle Jet, too.

Thanks shooter. There're a heck of a lot of details to keep up with, aren't there?

I tried to graph the Hammond 290HAX for the 5AR4. A bit lower plate voltage I think.

[dwinstonwood]

So, let's say 9.8/355=27.6ma + 2ma screen current = 29.6 x 2 = 59.2 + 6ma 12AX7 current = 65.2ma. I tried to draw the graph at 65ma and 275V.

[shooter]

There're a heck of a lot of details to keep up with, aren't there?

 
that's why humans invented computers 

[66Strat]

Cool. But, I'm going to try and stay the course with the 6G3. In that first video, near the end, sounded like X's Billy Zoom playing Hungry Wolf. Maybe I need a Sparkle Jet, too.

Thanks shooter. There're a heck of a lot of details to keep up with, aren't there?

I tried to graph the Hammond 290HAX for the 5AR4. A bit lower plate voltage I think.

Good estimate.

The transformer is rated 275V at 100 ma. The no load voltage is 296.4 volts per plate. The transformer AC voltage will sag down from 296.4 volts at 0 ohms current draw to 275 volts at 100 ma. With a 65 ma. load the AC voltage will be a little higher than 275 volts ~282 volts. Not enough difference to matter, just mentioned to show the effect transformer IR losses have. The lower the VAC current rating, the more the transformer voltage will sag under load.

[dwinstonwood]

Thank's for teaching me how to graph all of this! It's a great thing that this can be done visually on a graph. I need to be able to "see" things to really understand them.
I've heard the amplifier term "sag" for years and only thought of it as a tone factor that some might be able to hear from the speaker. But now I can see what it actually is, why it happens, and how it affects the operation of the power tubes.

And, 282V - ironically - graphs out closer to the 365 plate voltage on the schematic. Makes me think Fender might have been using the same PT when originally designing the 6G3, 5F10, and 5E11...

The Harvard 290HAX is a bit more expensive, but it's a much better match than the 290BX that Hammond reccomends for the 6G3. The 290BX seems more like an AB763 PT.

Well, I've burnt up a few brain resistors trudging through all this, but I'm a lot farther along than I was. Much appreciated 66Strat.

[66Strat]

Thank's for teaching me how to graph all of this! It's a great thing that this can be done visually on a graph. I need to be able to "see" things to really understand them.
I've heard the amplifier term "sag" for years and only thought of it as a tone factor that some might be able to hear from the speaker. But now I can see what it actually is, why it happens, and how it affects the operation of the power tubes.

And, 282V - ironically - graphs out closer to the 365 plate voltage on the schematic. Makes me think Fender might have been using the same PT when originally designing the 6G3, 5F10, and 5E11...

The Harvard 290HAX is a bit more expensive, but it's a much better match than the 290BX that Hammond reccomends for the 6G3. The 290BX seems more like an AB763 PT.

Well, I've burnt up a few brain resistors trudging through all this, but I'm a lot farther along than I was. Much appreciated 66Strat.

Very possible. As I mentioned before, the 375 volt number looked like an artifact from a previous iteration. It's also close to the voltage that would be expected with a 5Y3 rectifier under load at 333 VAC.

[dwinstonwood]

One manufacturer I didn't really look into is Heyboer. They have no specs on their website, and since I wasn't sure what specs I needed I didn't bother to email them. But, I'm considering the PT issue resolved. I'll order a 290HAX.

Also, I might just go ahead and order a custom Hoffman turret board instead of the eyelet version. I already have experience soldering up turrets, and I know what I'll do differently/better this time. Like following these instructions more closely:
https://el34world.com/Hoffman/instructions.htm

This drawing has the same dimensions as the eyelet board and follows the Fender layout as close as I get it.
Still a lot of drawing left to do...

[bmccowan]

Home again. That Hammond should do fine.

It was a ClassicTone PT - and I remember being confused by the spec sheets  I do not recall which model, but I'll look when I get home, as I'm away for the weekend. I think it was one of the ones with 2 secondary taps and I used the lower voltage red/white taps. But can't trust the memory. I was thinking I built it in 2017, but I found the chassis order and it was 2013!

As is typical my memory was wrong. The PT is a Weber 25130 which is 540V@150ma on the red/white lower voltage secondaries. I have just a tick under 350 on the plates using a GZ-34. I have not used Weber transformers for quite a while, not because they have failed me, but I just prefer to buy US made components. I did use a classic tone OT. If you are going to go the adjustable bias route, the Weber Lenora has a nice schematic and layout of that arrangement for a 6G3 circuit. I had to monkey with the value of the trim resister as the stock value didn't allow me to get to the sweet spot.

[dwinstonwood]

Thanks bmccowan! I grabbed the Lenora schematic and layout from the Weber site. The designer said he played with some of the power supply voltage divider resistor values to get the plate voltages on his schematic. It will be interesting to see where I land with the Harvard PT, and whether or not I should adjust anything from the original schematic. But, the bias pot will definitely go in the amp! Much appreciated.

I need to start buying parts (I put together a BoM in a Google spreadsheet), because I keep getting tempted to make modifications. For, example: grafting on a 6GXX-like preamp with treble and bass, and/or getting in way over my head by swapping in the harmonic vibrato circuit. Or, building a 6G11 Vibrolux with 6V6GT's, and so on...

[shooter]

need to start buying parts

buy them by 10's on the commons, makes the next build cheaper

[dwinstonwood]

buy them by 10's on the commons, makes the next build cheaper

You don't even have to say it. I paid more on shipping fees with the AC4 than I did for everything on the turret board.

[bmccowan]

I grabbed the Lenora schematic and layout from the Weber site. The designer said he played with some of the power supply voltage divider resistor values to get the plate voltages on his schematic. It will be interesting to see where I land with the Harvard PT, and whether or not I should adjust anything from the original schematic.

There is also a downloadable BOM spreadsheet for all of the Weber amps. Of course it has Weber part numbers but also the generic description. It can be edited so comes in handy as a shortcut for a BOM. I'm sure there are other BOMs out there too.
IMO, don't sweat hitting the exact B+ numbers too much. I sometimes play this amp with a 5R4 (not 5AR4) which adds sag dropping the B+ close to 40V. Amp still sounds great but somewhat different; you get nice grind at low volumes; kind of an oldtime blues vibe.     

[dwinstonwood]

Thanks bmccowan. Yeah, I'm not focusing too much on the schematic plate numbers anymore. 66Strat and you were finally able to get me to see that staying near or under 70% dissipation is the important target to hit.

One question I do have concerns resistors. I only selected 3 watt resistors for plate load, the B+ power filter section, and cathode bias locations. Are 1/2 watts fine for everywhere else, or is there another place where I need higher wattage rated parts?

Thanks

[bmccowan]

Well this is controversial - resistor sniffing and capacitor sniffing. So I will tell you what I do - but there are many other opinions.
I rarely use carbon comp, unless I'm repairing a valuable old amp.
For most locations I use either 1/2 watt metal film - or 1 watt carbon film. This is partly due to Sluckey's warnings about the lower quality of 1/2 watt carbon films.
For plate load I now use Dale military spec metal films - Tube Depot and the big houses like Mouser have them - the W rating is confusing as the military rates these differently. But they are low noise (not that I can really tell, but what the Hell, they have values printed on them instead of color bands!)
For power rail - I think 3 & 2 watt are fine for amps in this category, but I usually use a 5 or 10W ceramic in the first dropping position.
There is not need to go any higher than 1/2 watt for other resistors including preamp cathode, plate load. Some go higher stating that higher rating resisters create less noise. Well, my ears are hammered, so I can't hear any difference.
Now - on to the signal cap sniffing - just kidding.

[dwinstonwood]

OK, I hope I can explain this question in the correct way:

On the original schematic the bias voltage was tapped from a 333V AC secondary lead. The bias supply circuit used a 100K dropping resistor to get to -26V. See detail image.

The PT I intend to buy has a dedicated 50V AC secondary bias voltage supply. I modeled this in LTSpice. In order to achieve the -26V bias voltage I had to change the 100K resistor to a 3.9K resistor. This gave me a negative voltage very close to -26V. See detail image.

My question is: Have I successfully estimated a good dropping resistor value to start with given the 290HAX's 50V AC secondary bias supply?

Thanks!

[bmccowan]

I don't model, but a few things.
You are right that you do not want a 100K resistor. That is being use to knock down the B+ to a useful range for bias voltage. I believe I used a 10K for that resistor. But you plan to use a bias trim pot, as I did, eh? If that's right, take a look at that Lenora schematic as its a 6g3 with a 40-50v PT bias tap and a bias trim pot. I recall playing with the resistor values to get the trim pot in a useful range. Keep in mind that it can vary with the pair of power tubes you use. Others can help you with the math and modelling - I tend to see what others did and then; trial>error>error>success. Both approaches reach the same destination.

[dwinstonwood]

Thanks bmccowan. I don't put too much faith in LTSpice because I don't know it well enough to know if I'm using it right. I did see the 10K in the Lenora schematic and I started with that, and ended up with the odd 3.9K. Like you, I will more than likely have to try a few resistor values until it's right. But, all of this is a ways down the road.

[PRR]

> I don't put too much faith in LTSpice because I don't know it well enough to know if I'm using it right.

So what is "SINE 50 120"?

50V 120Hz? 50Hz 120V?

In my spice, the simmed V is *peak*. Transformers are marked in RMS. So if 50V transformer, my spice wants "70.7V".

120Hz wall-power is unlikely. Yes, a FULL-wave working on 60Hz makes 120Hz, but this is a halfwave; anyway you are specifying the "source" not the "result".

[dwinstonwood]

> I don't put too much faith in LTSpice because I don't know it well enough to know if I'm using it right.

So what is "SINE 50 120"?

50V 120Hz? 50Hz 120V?

In my spice, the simmed V is *peak*. Transformers are marked in RMS. So if 50V transformer, my spice wants "70.7V".

120Hz wall-power is unlikely. Yes, a FULL-wave working on 60Hz makes 120Hz, but this is a halfwave; anyway you are specifying the "source" not the "result".

Once again PRR, you catch my mistakes! Thanks

I completely forgot about the 1.414 multiplier. I had learned about that when I first started using LTSpice, and then I immediately proceeded to un-learn it.

Correcting the AC voltage to 70.7 peak and going back to a 10K resistor results in...
about -25VDC.

[66Strat]

Hoffman has a nice write up on adding a bias pot if you choose to do so. I think that a bias pot combined with 1 ohm 1% cathode resistors to provide tube current measuring points would make bias adjustment a lot easier.

https://el34world.com/charts/Biascircuits.htm

[dwinstonwood]

Thanks 66Strat. Yep, the 1ohm resistor method is definitely something I'll do. My hands aren't steady enough, and my nerves not brave enough to measure across the OPT primaries, or, whatever the other method entails.

I'll read the Hoffman guide.

[dwinstonwood]

BTW, it's apparently OK to mod 6G3's:

https://reverb.com/item/8002544-dumble-modded-fender-brown-deluxe-1962?utm_source=google&utm_medium=cpc&utm_campaign=6481783183&merchant_id=240050081&utm_content=813512089571&gclid=EAIaIQobChMIks3Kvs6d6gIViq_ICh1dWQ6BEAQYASABEgInPvD_BwE

[dwinstonwood]

I'm not sure how this will actually work out in real life, but in my model I need a 10K bias pot (which Hoffman sells), and the resistor values in the image below.
Doing this LTSpice stuff at least helps me see how the circuit works and how the components interact with each other, so it's worth my time.

[bmccowan]

Howard Dumble modded 6G3 - $22,100.  But the seller is accepting offers. Maybe he'd knock the 100 bucks off?

[shooter]

 
push it;
tell the computer to predict the voltage range -23 to -31 kinda thing

[sluckey]

   It's not a sin to tape off that bias tap and use the HT winding just like the original amp.

[HotBluePlates]

... 66Strat, thanks for the starting point calculations. The 270EX/5AR4 combo looks good to me.

The one blindspot for me is how you arrived at 65ma? Why not 60ma and 370V, for example. You mentioned "28.85 ma current for each 6V6 tube." What am I missing here?

If you look at the 6G3 schematic, it shows a plate voltage of 365 volts. ...

Just as with the advice to not put too much faith in LTSpice, don't put too much faith in the voltages listed on the 6G3 Deluxe schematic.

I've got an original 1962 Deluxe with its original 125P17A power transformer (Fender used 3 different PTs in this amp over time), and was told by other 6G3 owners that my amp's PT delivered lower B+ voltages than the 125P2A.

      - My amp lands on 6.3vac for the heater winding at a hair over 117vac on the primary.

      - When fed 117vac, I have 411vdc at the 1st filter cap, 405vdc on the 6V6 plates and 402vdc on the screens when using a GZ34 and drawing 24.7mA (~71% of 14 watts).  Bias was at -33vdc.

      - The high voltage winding output is 326.5v - 0 - 326.5v in this configuration.

      - Preamp supply & plate voltages throughout are ~30v higher than the schematic says.

I used to also own a 1963 6G3 Deluxe and the 1st filter cap B+ was 440vdc, with 433vdc at the 6V6 plate, 432vdc screen, and bias voltage at -32.5vdc.  Don't have the 6V6 plate current in my notes for this amp, but the power transformer was the third type as used in the Deluxe Reverb amps that followed the next year.

So don't get too hung up on the exact voltages listed in the schematic, and do add a bias pot to the bias supply.  It just makes everything easier.