Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Blooze on July 14, 2022, 02:51:39 pm

Title: Vibrochamp biasing
Post by: Blooze on July 14, 2022, 02:51:39 pm
I recently went through my SF Vibrochamp and recapped it. I replaced the can cap with a 20/20/20/20 and put a 1K/10W resistor between the first and 2nd 20uF to bring the voltages down before going to the plate. I also added a 1K/3W screen resistor based on a suggestion from several folks. Using a 680R cathode resistor.

Here are the voltages I'm now getting: Attached a schematic too.
From the 5Y3: 419V
After the 1K dropping resistor: 378V
Plate: 367V
Cathode: 24.8V
Screen: 264V (6V drop across the screen resistor)

My question is if I'm dropping 6V (270V to 264V) across that 1K screen resistor doesn't that take my tube dissipation from 12.5W down to 10.4W when the screen current is figured in? That's about a 16.5% screen current, correct.

I know this amp was running at about 18W dissipation originally, but it also would torch a new production 6V6 is short order. That's why I dropped the voltage and bias current. I just never actually figured in the screen current to the bias equation.

So should i just pull the screen resistor, or try lowering it to 470R or so? Lower the cathode resistor back to 470R?  I I intially left the 470R cathode resistor in and was getting about 15W with the lowered B+.
Title: Re: Vibrochamp biasing
Post by: HotBluePlates on July 14, 2022, 03:46:51 pm
Quote from: Blooze on July 14, 2022, 02:51:39 pm
... doesn't that take my tube dissipation from 12.5W down to 10.4W when the screen current is figured in? ...

Yes.

Quote from: Blooze on July 14, 2022, 02:51:39 pm
I know this amp was running at about 18W dissipation originally, but it also would torch a new production 6V6 is short order. That's why I dropped the voltage and bias current. I just never actually figured in the screen current to the bias equation.

So should i just pull the screen resistor, or try lowering it to 470R or so? ...

I'd argue you should use a 470Ω cathode resistor.  But then also look at the resistor feeding the filter cap connected to your screen resistor:  stock is 1kΩ, try 18kΩ (2 or 3 watt).  Consider having 10kΩ, 15kΩ, 22kΩ as backups.


Why?

You will be dropping the screen voltage (and in a better way than just using a stouter series screen resistor).  The lower screen volts will reduce plate current.  That will make it so you do not need as-large a cathode resistor, and the amp will seem to have better "power sensitivity."
Title: Re: Vibrochamp biasing
Post by: Blooze on July 14, 2022, 04:07:02 pm
Quote from: HotBluePlates on July 14, 2022, 03:46:51 pm

Quote from: Blooze on July 14, 2022, 02:51:39 pm
I know this amp was running at about 18W dissipation originally, but it also would torch a new production 6V6 is short order. That's why I dropped the voltage and bias current. I just never actually figured in the screen current to the bias equation.

So should i just pull the screen resistor, or try lowering it to 470R or so? ...

I'd argue you should use a 470Ω cathode resistor.  But then also look at the resistor feeding the filter cap connected to your screen resistor:  stock is 1kΩ, try 18kΩ (2 or 3 watt).  Consider having 10kΩ, 15kΩ, 22kΩ as backups.


Why?

You will be dropping the screen voltage (and in a better way than just using a stouter series screen resistor).  The lower screen volts will reduce plate current.  That will make it so you do not need as-large a cathode resistor, and the amp will seem to have better "power sensitivity."

Won't that end up effecting the 12AX7's by dropping the voltage after the 3rd filter? Or would I just then adjust the 10K between the 3rd and 4th filter cap? The 12AX7's should self bias anyway though, correct.
Title: Re: Vibrochamp biasing
Post by: shooter on July 14, 2022, 05:50:58 pm
Quote
I recently went through my SF Vibrochamp


how close is the old sound to the new sound?


IF you're aiming for the old sound, i would think tube abuse would be a small value in the real math
IF you're aiming for something new, roll a bunch of Rs through the PA section


Quote
Won't that end up effecting the 12AX7's


see above argument then match the old values, or...
Title: Re: Vibrochamp biasing
Post by: Blooze on July 14, 2022, 05:57:51 pm
Quote from: shooter on July 14, 2022, 05:50:58 pm
Quote
I recently went through my SF Vibrochamp


how close is the old sound to the new sound?


IF you're aiming for the old sound, i would think tube abuse would be a small value in the real math
IF you're aiming for something new, roll a bunch of Rs through the PA section


Quote
Won't that end up effecting the 12AX7's


see above argument then match the old values, or...

Not a lot of difference in sound that I can tell. At the original values it would chew up a new 6V6 fairly quickly. JJ’s have a 14W max. As expensive as tubes are getting, and it’s just gonna get worse, I figured if I could get it to 12W it’d be better than the 18-19W I started at.
Title: Re: Vibrochamp biasing
Post by: Blooze on July 14, 2022, 07:01:51 pm
Quote from: HotBluePlates on July 14, 2022, 03:46:51 pm

I'd argue you should use a 470Ω cathode resistor.  But then also look at the resistor feeding the filter cap connected to your screen resistor:  stock is 1kΩ, try 18kΩ (2 or 3 watt).  Consider having 10kΩ, 15kΩ, 22kΩ as backups.


Why?

You will be dropping the screen voltage (and in a better way than just using a stouter series screen resistor).  The lower screen volts will reduce plate current.  That will make it so you do

You’re basically describing this, correct? It makes sense.



So go back to a 470R cathode resistor. Change the 1K screen supply resistor to a 10K-22K. Since I’ve already knocked my B+ down from 420V to the ~380V I can basically just change that screen resistor. Pull out the 1K series screen resistor I installed on the tube socket?

What screen voltages should I be shooting for. ~300V?  If my B+ stays at 375 or so at the plate?  I don’t have a signal generator so I can’t really measure the current change at full volume.

Thanks for the help!

Edit: I found your explanation over at TGP and why 18K would be a good starting point.
Title: Re: Vibrochamp biasing
Post by: dude on July 14, 2022, 09:01:00 pm
Quote from: Blooze on July 14, 2022, 07:01:51 pm

Edit: I found your explanation over at TGP and why 18K would be a good starting point.
What is TGP..?
Title: Re: Vibrochamp biasing
Post by: Blooze on July 14, 2022, 09:21:38 pm
Quote from: dude on July 14, 2022, 09:01:00 pm
Quote from: Blooze on July 14, 2022, 07:01:51 pm

Edit: I found your explanation over at TGP and why 18K would be a good starting point.
What is TGP..?


The Gear Page.

https://www.thegearpage.net/board/index.php?threads/d-lab-electronics-makes-a-fender-vibro-champ-sound-better-than-new.2343139/
Title: Re: Vibrochamp biasing
Post by: HotBluePlates on July 15, 2022, 05:49:37 am
Quote from: Blooze on July 14, 2022, 04:07:02 pm
Won't that end up effecting the 12AX7's by dropping the voltage after the 3rd filter? ...

I think you have it figured out by now with the stuff you found (and the comments you made with your question were spot-on).
Title: Re: Vibrochamp biasing
Post by: dude on July 15, 2022, 10:58:25 am
Blooze have you tried HotBluesPlate’s suggestions, dropping R’s, 10k to 18k, first node? And any other higher valves down the line. Did you keep the 470k cathode R, scrap the screen grid R..?
Let us know what you did and the resulting tone.
I too have a 67 Vibro Champ and just raised the cathode R to 680 and added a 470 screen, maybe l should try HBP’s suggestions.
Title: Re: Vibrochamp biasing
Post by: tubeswell on July 15, 2022, 11:50:33 am
Another vote for 470R cathode resistor, and 10k screen supply dropping resistor. (In one of these amps, running the sorts of B+ it runs, this keeps the 6V6 screen voltage about 50v below the plate voltage, while running the tube at Pmax. This results in a load line that hits just below the knee of Vg0 grid curve and ensures screen current isn’t too excessive when the amp is pushed hard.) Look at a schematic for a 5F2A narrow panel tweed Princeton - set up the same way and sounds awesome.


The preamp triodes won’t mind working at the (50V) lower B+ voltage either. Having a higher B+ isn’t as important for their sound (in one of these amps), as having the plate idling at 2/3 of HT with a load resistance = about 2 x the plate resistance.

Title: Re: Vibrochamp biasing
Post by: dude on July 15, 2022, 12:23:35 pm
Tubeswell, so you too recommend notes A & B, 10k and 22k, no screen R, 470 cathode R. And don’t worry about the voltage drop to the preamp.
I love the sound of a 5F2A Princeton, which is exactly as above.
I wasn’t exactly happy with the tone of my Vibro Champ, l am running a 10” spk but l did change that tiny OT, for an 8 ohm, 10 watt SE, old paper wound 40:1 OT
EDIT: 34:1 ratio
Title: Re: Vibrochamp biasing
Post by: tubeswell on July 16, 2022, 12:21:32 am
Quote from: dude on July 15, 2022, 12:23:35 pm
I did change that tiny OT, for an 8 ohm, 10 watt SE, old paper wound 40:1 OT
EDIT: 34:1 ratio


You can quite often find an old Stancor or Thordarson 5-10W SE OT on eBay. These older OTs were designed to take more 'one-way' magnetising current without saturating, but are not excessively big so they run just under the edge of saturation, and they sound really awesome in a 5W SE 6V6 amp. I think the key to a great SE amp sound is to ensure the OT isn't too big (because too much iron makes it too clean and kills the mojo) YMMV
Title: Re: Vibrochamp biasing
Post by: Blooze on July 16, 2022, 09:17:53 am
Quote from: dude on July 15, 2022, 10:58:25 am
Blooze have you tried HotBluesPlate’s suggestions, dropping R’s, 10k to 18k, first node? And any other higher valves down the line. Did you keep the 470k cathode R, scrap the screen grid R..?
Let us know what you did and the resulting tone.
I too have a 67 Vibro Champ and just raised the cathode R to 680 and added a 470 screen, maybe l should try HBP’s suggestions.

I’ll need to order some more resistors. I thought I had some 15k and 22k, but nope. Mines a ‘76 Silverface.

I plan on removing the series screen resistor and putting a 470R back in the cathode. I’ll probably start with the 18K that was recommended.

I’m debating removing, leaving, or dropping to 470R the B+ dropping resistor I added between the extra first and second filter cap of my 20/20/20/20 can cap.
Title: Re: Vibrochamp biasing
Post by: pdf64 on July 16, 2022, 09:28:49 am
Quote from: Blooze on July 16, 2022, 09:17:53 am
…
I’m debating removing, leaving, or dropping to 470R the B+ dropping resistor I added between the extra first and second filter cap of my 20/20/20/20 can cap.
So its OT is fed from the 2nd filter cap?
That’s doubly beneficial, as it drops HT voltage to the 6V6, and filters out obnoxious 120Hz hum.
What’s the argument for removing it?
Title: Re: Vibrochamp biasing
Post by: dude on July 16, 2022, 10:13:42 am
The schematic Blooze posted shows a 1k after 5y3, l assume he replaced it with a 470….? That 470 is not dropping much voltage, maybe the 1k would be better if you decide to use it.
I have a BF Vibro Champ and the only difference l see from the Silver is a bigger PT voltage and 4 20uf’s vs 3 in the BF.
Since you can drop the plate and screen voltage with the two dropping R’s, you don’t need that 470 or 1k, unless you have hum or excessive noise as pdf64 suggested.
Another option would be to tie the first two 20uf’s in parallel for 40uf, as in several early Fenders like the Harvard. Might produce a tighter bass response.
In my BF, l have a 33uf, then two 22ufs, with the lower voltage in the BF PT, l used a 15k and 10k dropping resister, kept the 470 biasing cathode R and no screen. Of course you may want to change these value’s as you have more voltage to start with. I finished the changes but cleaning up some sloppy work, so l haven’t tried it yet.
But the changes in tone between the small champ OT vs the old paper wound 8/10 watt l mentioned above was fantastic, that small OT saturates very fast and the 8/10 watt gives a little more headroom and still breaks up like the original. And those old paper wound SE OT’s have a secondary of around 7000 ohm load resistance, great for a 10” 8 ohm speaker which l have.
I didn’t butcher my Champ, used a new baffle and grill, kept the original with the 8” speaker, kept the original OT and can always put it back.

Title: Re: Vibrochamp biasing
Post by: Blooze on July 16, 2022, 11:00:31 am
The schematic I posted is how it sits currently. 1K on that dropping resistor. Dropped my plate voltage about 25 volts. My original can cap was a 40/20/20.

Title: Re: Vibrochamp biasing
Post by: dude on July 16, 2022, 12:12:25 pm
I’d leave it like that schematic, like pdf64 thumbs up, that cap is filtering the 5y3 and lowering voltage, a good thing. Just do the changes on the other two dropping R’s, like maybe 18k and 10k, to start.  If you like a tweed tone, you could try a 22k to the preamp instead of 10k. You could make a resistor jumper and try different values, see what you like.  Your video was interesting, thank you for posting. I was old school, changing the cathode R to lower screen voltage, low volume amp sounded nice but turn up not so, blocking distortion l believe. I’ll post here the results: Thanks again for that video.
Title: Re: Vibrochamp biasing
Post by: pullshocks on July 25, 2022, 05:43:39 pm
It seems the OP has no further questions, so pardon me if I partially hijack the thread.
 This thread has got me thinking about my 6V6 SE amp that I built a number of years ago.  It has had a few different preamps and tube effects loops in it, but I haven’t really paid any attention to the power amp.  Right now I am calling it the “Champerial” based on the Tone King Imperial clean channel and the 6V6 SE PA.  After reading this tread I realize this doesn’t make sense because the power amp has a number of differences compared to the Champ power amp.
It mostly follows the AX 84 5W SE “building block” but with a higher voltage PT.  I used a tube rectifier to compensate, and ended up with slightly lower B+ voltages (and much lower than Champ voltages).
The AX84 power amp has some differences compared to the Champ circuits- extra power supply node between the standby switch and OT B+ connection, lower voltages, 56K grid resistor, no 220K to ground on the 6V6 grid, 1K screen resistor, different dropping resistor values, and no NFB. 
Today I mapped out the as built schematic, and measured voltages, screen current, and cathode current. I see that I deviated from the AX84 power supply circuit by positioning the choke between the 2nd and 3rd nodes instead of between the 1st and 2nd.  Not sure why I did that, either an error or something I read on the internet at the time about reducing hum in SE amps, or ???
Either way, it has worked and sounded fine for 15+ years.  But would it be better to run the choke first?
After tweaking the adjustable cathode bias circuit to the AX 84 spec, i.e. cathode current=.9x max plate dissipation/plate voltage (Specifically .9x14/302=41.8mA), at idle I get
Plate voltage 302
Screen voltage 298
Screen current 3.3 mA
Cathode voltage 19.6
Cathode current 41.8 mA
The total cathode resistance is 451 ohms.
Any red flags here?
Other than increased power (which I don’t need) would there be any advantage in increasing voltages via a 5V4  or 5AR4 rectifier tube and/or changing to a topology like the champ circuit recommendations discussed above ?
Thanks in advance
Pullshocks/Mark
Title: Re: Vibrochamp biasing
Post by: PRR on July 26, 2022, 12:40:52 am
> between the 2nd and 3rd nodes instead of between the 1st and 2nd.

Don't go by numbers. Your choke is between plate and screen. Here is only needs to be 6mA instead of 47mA, a neat saving. If the amp makes you happy, be happy.
Title: Re: Vibrochamp biasing
Post by: pullshocks on July 26, 2022, 02:49:02 am
Quote from: PRR on July 26, 2022, 12:40:52 am
> between the 2nd and 3rd nodes instead of between the 1st and 2nd.

Don't go by numbers. Your choke is between plate and screen. Here is only needs to be 6mA instead of 47mA, a neat saving. If the amp makes you happy, be happy.


Thanks PRR, this is the choke I used Choke - Hammond, for Fender Deluxe, Deluxe Reverb, etc. | Antique Electronic Supply (tubesandmore.com) (https://www.tubesandmore.com/products/choke-hammond-fender-deluxe-deluxe-reverb-etc).  The current rating is 47 mA, less than the AX84 spec (70 mA), but I gather the choke in the Champ 5E-1 is rated at 47 mA...


Yes I am happy enough with the amp in its current form, but I'm always on the lookout for circuit changes that would make me even happier.  Its the "grass is greener on the other side of the fence" thing that has kept me reading tube amp forums, and building for almost 30 years. 



Title: Re: Vibrochamp biasing
Post by: shooter on July 26, 2022, 04:22:42 am
Quote
"grass is greener on the other side of the fence"


 :laugh: ....til you jump the fence and find the grass is green from fresh aromatic cow dung
Title: Re: Vibrochamp biasing
Post by: ac427v on July 26, 2022, 07:28:18 am
Pullshocks already has a 100 ohm/10watt resistor between two reservoir caps feeding the 6V6 plate. I doubt if there would be any audible difference if he replaced that resistor with a choke. I like that design. It would be easy to try though...
Title: Re: Vibrochamp biasing
Post by: jordan86 on July 26, 2022, 07:48:12 am
If I may insert a “noob” question, you all are discussing adjusting the bias/tone via 3 different means

1) Adjusting overall B+/Plate voltage
2) Adjusting the cathode bias resistor
3) Adjusting screen resistor value?

That last one is mysterious to me. As I understand it, the screen voltage can effect the efficiency of the plate, allowing it to produce full power potential, or not. Or are you saying that a screen resistor arrangement actually changes the plate dissipation as well.

That’s foreign territory to me, but I’d be super interested to hear more. Sounds like a worthy deep dive. Where can I read up on that?
Title: Re: Vibrochamp biasing
Post by: 66Strat on July 26, 2022, 09:35:46 am
Plate current is controlled by both the control grid and the screen grid. As screen grid voltage increases, plate current increases. The graph at the top of page 5 of the attached illustrates the relationship of plate current to changes in screen voltage and control grid bias voltage.

http://www.r-type.org/pdfs/6v6gt.pdf (http://www.r-type.org/pdfs/6v6gt.pdf)
Title: Re: Vibrochamp biasing
Post by: dude on July 26, 2022, 11:54:58 am
The original post was about biasing. Having a much lower screen grid voltage (about 50vdc less) than the plate on the 6V6, was it not...? About biasing the 6V6, members decided to raise the dropping resister to the screen lowering voltage about 50v's less than the plate, as the original poster was eating 6V6's. Lowering the screen node R, rather than raising the cathode R to bias cooler...? 
Title: Re: Vibrochamp biasing
Post by: dude on July 26, 2022, 07:37:49 pm
Quote from: pullshocks on July 25, 2022, 05:43:39 pm
It seems the OP has no further questions, so pardon me if I partially hijack the thread
Pullshocks, you didn’t hijack the thread, l apologize for wording my reply as if you did, if I did.  Perfectly normal to ask your question about SE 6v6 amp.  :BangHead:
Title: Re: Vibrochamp biasing
Post by: pullshocks on July 26, 2022, 11:36:27 pm
Quote from: dude on July 26, 2022, 07:37:49 pm
Quote from: pullshocks on July 25, 2022, 05:43:39 pm
It seems the OP has no further questions, so pardon me if I partially hijack the thread
Pullshocks, you didn’t hijack the thread, l apologize for wording my reply as if you did, if I did.  Perfectly normal to ask your question about SE 6v6 amp.  :BangHead:


Dude, not at all.  My reaction was that you gave a concise summary.  So from my standpoint no apology needed, but I appreciate the thought.


If I understand the thread, at vibrochamp voltages, dropping the screen voltage is needed, and dropping resistors are a more preferable way to do this than screen resistors--as PRR says, " ...the amp will have better "power sensitivity.""


So at the lower voltage in my amp, would the same apply?  Currently, plate voltage is 302 and screen 299. 

Title: Re: Vibrochamp biasing
Post by: HotBluePlates on July 30, 2022, 06:46:30 am
Quote from: jordan86 on July 26, 2022, 07:48:12 am
If I may insert a “noob” question, you all are discussing adjusting the bias ... via 3 different means

1) Adjusting overall B+/Plate voltage
2) Adjusting the cathode bias resistor
3) Adjusting screen resistor value?

That last one is mysterious to me. ... are you saying that ...

Blooze linked to a thread at another forum, but you might not have read far enough to find it.  Check out this post (https://www.thegearpage.net/board/index.php?threads/d-lab-electronics-makes-a-fender-vibro-champ-sound-better-than-new.2343139/post-34379239).


People notice their Vibro Champs have excessive dissipation, and nearly double the cathode resistor to address the issue.  I contend that is not desirable because the resulting large voltage (from G1 to cathode) makes the output tube less sensitive.  So I offered the alternative of lowering screen voltage to reduce plate current, keeping the same small cathode resistor (and small voltage G1-to-cathode), and so the output tube is more-sensitive.

"Sensitivity" here is how much drive signal is required to get the output tube to make its max clean output power, and then distort.  So higher sensitivity means the same preamp has an easier time driving the output tube to distortion.

A primary complaint of those who simply used a large cathode resistance to tame idle current was their amp won't distort even when they turn it full-up.  Knocking down screen volts (in the power supply) addresses both issues.
Title: Re: Vibrochamp biasing
Post by: pullshocks on July 30, 2022, 03:43:19 pm
66Strat thanks for posting the tube data.


Interesting to see what they have for plate and screen voltages for "typical operation"
Title: Re: Vibrochamp biasing
Post by: HotBluePlates on July 30, 2022, 06:59:27 pm
Quote from: pullshocks on July 30, 2022, 03:43:19 pm
Interesting to see what they have for plate and screen voltages for "typical operation"

"Class A Amplifier" translates to "Single Ended."

A table radio would have been the prime spot to use a single-ended 6V6.

If you're like many folks down-sizing from a 20-50 watt amp, you might have thought "4 watts will be quiet."  And like many of them, you may have been surprised to find a 4 watt Champ is still way-loud when it distorts.  It really takes around 1/10th watt to be "TV loud" in an typical guitar speaker.



Whether that table radio makes 3 watts or bumps up the voltage to make 4 watts barely matters when "1/2 watt" was subjectively "way loud."

There's also a historical connection to battery voltages driving the figures found on tube data sheets.
Title: Re: Vibrochamp biasing
Post by: PRR on July 30, 2022, 09:55:03 pm
> also a historical connection

6V6 suggestions come from 6F6 suggestions, because many 6V6 went into existing 6F6 designs.
Title: Re: Vibrochamp biasing
Post by: Blooze on September 01, 2022, 01:59:46 pm
Ok folks. I replaced the 10K CC resistor with a 10k metal oxide, the 1K CC with an 18K. Put a 470R back into the the 6V6 cathode and removed my 1K screen resistor. Voltage to the 5Y3 is 375V.

6V6 plate is 376V. Voltage across 470R cathode resistor is 15.04V.  Pin 4 screen to 6V6 is 285V. Voltage to OPT is 384V. So that’s about 11.5W or so.

Amp starts to breakup about 4 on the volume and just gets more glorious. Way better than before.

Title: Re: Vibrochamp biasing
Post by: Blooze on September 01, 2022, 03:25:08 pm
Threw in a new JJ 6V6 to see the difference from the Raytheon 6V6gt and it’s about 11.7W. So no problem going to the new tubes as well when I need to.
Title: Re: Vibrochamp biasing
Post by: dude on September 01, 2022, 04:02:33 pm
You can also try raising the 10k to preamp, say to 18/22k 3W. This change will put you in Champ Tweed territory, or Brown/tone. You may or may not like the brown tone, lower preamp voltages. You can lift one leg, tack in another 10k in series, just to test or use a jumper.
Title: Re: Vibrochamp biasing
Post by: HotBluePlates on September 02, 2022, 06:25:59 pm
Quote from: Blooze on September 01, 2022, 01:59:46 pm
... I replaced ... the 1K CC with an 18K. Put a 470R back into the the 6V6 cathode ...

6V6 plate is 376V. Voltage across 470R cathode resistor is 15.04V.  Pin 4 screen to 6V6 is 285V. Voltage to OPT is 384V. So that’s about 11.5W or so.

Amp starts to breakup about 4 on the volume and just gets more glorious. Way better than before.

Nice!!

(https://pbs.twimg.com/media/DmVPPufU0AIu8mM.jpg)