6G3 Capacitor Question

[dwinstonwood]

Thanks for all the input everyone. I'm putting in at least one order tomorrow, so pics of real amp parts - instead of pc simulations - will appear soon.

sluckey, the thought of not using the 50V tap had crossed my mind. It's an alternative, for sure.

bmccowan, well, it does say "+$1 Shipping," so that could be considered a bargain.

shooter, I'm working on it!

Thanks HotBluePlates. Whatever my plate voltages end up at will be fine. I might be mistaken about this, but from what I understand higher plate voltages can provide more clean headroom. Wasn't there a Music Man 130, or something, that had like 700V on the plates for that reason? Leo Fender's ultimate clean amp?
Since this is for home use I'd probably be happier with plate voltages nearer 365, or less. That's really the main reason I settled on the 290HAX PT (275-0-275) instead of a Deluxe Reverb type. Your 326.5v - 0 - 326.5v secondary windings seem closer to the latter. So, correct me if I'm misunderstanding the effects of higher plate voltages. But, the bias pot and the 1 ohm resistors will definitely be added.

[dwinstonwood]

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

I took that as a challenge. And, you were pretty darn close to the voltage range I got.

(and, notice I read Hoffman's bias guide and used a 27k resistor and a 47uF cap)

[shooter]

per HBP;

Bias was at -33vdc.

now that you're a wizzz, see what R values you'll need to get it ~~~ -28 to -36 give or take a few 

[bmccowan]

Thank (whatever God you want) that tube amps are a mix of art and science. I can do the art part and fake the science part. Yes, you will be happy at any voltage within shouting distance of 350.

[dwinstonwood]

I just put in a pretty big order (for me, anyway) at Hoffman Amps. No turning back now. I did think to buy eyelet boards for the filter caps and bias circuit, along with the main board. I almost overlooked those.
AES has a 4th sale, so I'll save a bit on TX's and tubes. Mojotone has the chassis with front and back plates. I need to sell some stuff on eBay to justify a cabinet purchase. I'm still leaning towards a head. But, obviously, the amp can be built and tested without that.

Thank (whatever God you want) that tube amps are a mix of art and science...

bmccowan, fwiw, I'm pretty sure I failed every math class I ever took, and I know I never took a science class. I'm what they call an ABD art historian.

shooter, I was going to use Ohm's law, not the software, to solve that riddle (which bias adjustment resistor value to use for -33vdc). But, I hit a wall trying to figure out how to understand the resistances in the bias circuit. I can calculate current and voltage drops when the resistors are all in series, but this looks confusing to me because the 27k goes to ground, so I gave up. Maybe a 6k8 instead of 10k? The Harvard schematic shows that value...

[shooter]

but this looks confusing to me

that's why I build cathode biased amps 

[dwinstonwood]

but this looks confusing to me

that's why I build cathode biased amps 

Yep, I spent a few hours trying to figure out if it's even possible - or, necessary - to draw output tube load lines for a fixed bias amp, and totally gave up.

The parts, transformers, and chassis (with face plates) have been ordered. I know very little about tubes, so I need to do some research on which ones I'll want to buy. But, the Tung-Sol 6V6GT's look interesting.

bmccowan, did you make that head cabinet yourself? It doesn't look like anyone stocks a 6G3 head cabinet; I'd probably have to order something custom made. The 6G3's chassis length is unique. Thanks.

[sluckey]

I rarely calculate bias voltage resistor values. I find it more than adequate to just know if the negative bias voltage will increase or decrease if I change this or that resistor. Knowing this and having a full stock of one watt and half watt resistors gets me the proper bias voltage every time, usually within 5 minutes.

[bmccowan]

bmccowan, did you make that head cabinet yourself? It doesn't look like anyone stocks a 6G3 head cabinet; I'd probably have to order something custom made. The 6G3's chassis length is unique. Thanks.

Yes, I made it myself. I've been woodworking for quite a few years and have a full shop - so it was not difficult. You just have to careful to provide just enough clearance for the Tolex in the opening for the chassis. Finger joints are easy if you have a jig, a bitch if you don't. The Tolexing is not that difficult either, but I hate gluing that stuff on. Anyways, I think any of the companies that make that combo cabinet would make the head for you. Mather cabs look really nice. And I doubt Mojo makes their own cabs, so whoever does would likely make a head version of their cab to fit that chassis. There are also a few Ebay sellers who make nice bare pine cabs, if you want to do the covering. There are some really good tutorials on Youtube for gluing up the tolex or other covering.
On power tubes - I have a stash of NOS 6V6s, but I put JJs in my 6G3 and think they sound pretty darn good.

[dwinstonwood]

Thanks bmccowan. I actually wrote to Peter Mather today. We're in the same state, close enough to drive to, in fact! He's made 6G3 head cabinets for Jim Nickelson of Lil'Dawg in NV, whom I'm also communicating with, so I'm sure he can build what I want. BTW, Guitar Cabinets Direct and Mojotone are both in Burgaw, NC, and they offer the same cabinet products, so I'm sure they connected, maybe?

For the second half of the 80's I was a cabinetmaker in Philly. I made custom furniture. Currently, I don't have a shop or tools. I really miss woodworking, but I don't have the room, living in a townhouse in the city. Anyway, I would definitely build my own cabinets if I had the tools and the space.

[bmccowan]

Well you have the skills for sure. If I did not have a woodworking shop at home I would look into one of those shared shops (well post Covid I suppose) as I would really miss working with wood. At least amp building can be done in a more confined space. A friend of mine did a complete rebuild of a Harley Knucklehead in his living room - he's divorced. That story is sure to induce a Shooter story.

[shooter]

induce a Shooter story.

 

I'm in the home stretch, modernizing a 1930's barn foundation, my woodworking skills are rough 

[HotBluePlates]

per HBP;

Bias was at -33vdc.

now that you're a wizzz, see what R values you'll need to get it ~~~ -28 to -36 give or take a few 

... shooter, I was going to use Ohm's law, not the software, to solve that riddle (which bias adjustment resistor value to use for -33vdc). ...

You can't get a good answer using Ohm's Law for the bias supply (as drawn in the 6G3 schematic), because rectification is a non-linear process and doesn't adhere to Ohm's Law.

You could use Ohm's Law if you remove the 22kΩ resistor shown across the bias filter cap in the schematic, assume some voltage to be present at the schematic, and then replace the 22kΩ resistor with a rheostat+resistor.  However, as Sluckey said it's often easier to figure this out with a handful of resistors than with a pencil or software.

... I spent a few hours trying to figure out if it's even possible - or, necessary - to draw output tube load lines for a fixed bias amp, and totally gave up. ...

No need to bother with that.  You're not trying to design an amp, but rather build an amp to an existing design.  Your main challenge will be figuring out if the transformers & speaker used approximate the original amp's transformers & speaker, cause everything else is pretty well spelled out (use the same resistor values, but don't get too hung up on whether the amp voltage match the schematic voltages, as those don't seem to match the real amps anyway).

[66Strat]

Implement Hoffman's better bias circuit.
https://el34world.com/charts/Biascircuits.htm

You will need to play with resistor values for the circled resistor to get the adjustment range that you want. Once implemented, biasing the amp will be much easier anytime you change output tubes.

[dwinstonwood]

Thanks HotBluePlates. I've always been told I overthink things. If I encounter something that's not clear, it's in my nature to try and figure it out. Thanks for keeping me focused.

66Strat, I've been working on adding the Hoffman bias circuit to my layout. I've also been watching videos on measuring and adjusting fixed bias amps.

From what I understand, measuring across the 1 Ohm cathode resistors - that I will add - and measuring from pin 3 of the 6V6GT's to ground will give me the info I will need to calculate plate dissipation.
If it's not in the 60% to 70% ballpark, and I can't get it there with the bias pot, I will start changing the resistor you circled in red.

I'm guessing one needs to work quickly when doing this in case the negative grid (bias) voltage is way off.

I will measure the 290HAX's secondary bias tap, confirm that all resistor values are within tolerance, and quadruple check all wiring and solder joints before that first "roll-of-the-dice" start up.

[sluckey]

I'm guessing one needs to work quickly when doing this in case the negative grid (bias) voltage is way off.

The smart thing to do is set the bias pot for maximum negative voltage at pin 5 of the output tubes. DO THIS BEFORE YOU EVEN PLUG IN ANY TUBES!

I will measure the 290HAX's secondary bias tap, confirm that all resistor values are within tolerance, and quadruple check all wiring and solder joints before that first "roll-of-the-dice" start up.

I'm pretty sure that if you use the bias tap on that PT you will need to change that 10K to about 470Ω. Get the bias voltage right BEFORE YOU EVEN PLUG IN ANY TUBES!

Your head must be in overload from all this preliminary info. I hope you will remember some of this when/if you ever build this amp.   

[dwinstonwood]

Thanks sluckey. That's something that I think I can see and understand.

When I'm ready to take bias measurements I will:

1.) With amp off, rectifier in, but power tubes not installed, connect my test leads to pin 5 and ground. Power up and and rotate the bias pot until I have the most negative voltage. Check other tube socket, too.
2.) With amp off, rectifier in, but power tubes not installed, connect test leads to pin 3 and ground. Power up and measure plate voltage. Repeat for other tube socket.

But! Here's the blind spot I can't seem to see around:

If I know the grid negative voltage and the plate voltage, but not the voltage drop across the 1 Ohm cathode resistor, which I can't measure with the tube out, how do I calculate the current that will flow through the tubes? So, how will I calculate plate dissipation?

That's the catch that might be obvious, but I can't see it!

Thanks.

[sluckey]

1.) With amp off, rectifier in, but power tubes not installed, connect my test leads to pin 5 and ground. Power up and and rotate the bias pot until I have the most negative voltage. Check other tube socket, too.

HELLO. Are you listening? What did I say above? I even used all caps and colored it red.

If I know the grid negative voltage and the plate voltage, but not the voltage drop across the 1 Ohm cathode resistor, which I can't measure with the tube out, how do I calculate the current that will flow through the tubes? So, how will I calculate plate dissipation?

Don't worry about dissipation until you have the amp working and have heard a guitar played through it. Setting the bias is a fine tuning adjustment that comes later. Just be sure that you have at least -35V on pins 5 of the output tubes. But if you connect that PT bias tap to that 10K on the bias board you will not even be close to -35V or greater. Last time I'm gonna say this... Start with a 470Ω on that bias board.

This bias is such a simple circuit but it is so critical. Several expensive components rely on it to be correct and stable. It's OK to muck around with the bias circuit, but it's dangerous to muck about when you don't understand what you're doing. That's why early on I suggested to just build it according to the original schematic.

[dwinstonwood]

Sometimes it takes larger mallet.

Sorry, due to a serious brain lapse I pictured the bias supply exiting the rectifier. I know it comes straight off of the PT secondary, but confusion took over.

Thank sluckey. I'll use a 470Ω resistor.

...when/if you ever build this amp.   

All three of my orders have shipped. I should have them by the weekend!

Hey, thanks y'all for showing such a genuine interest in my project. I know I've made it a challenge for you.

[shooter]

can't recall if I passed this on, worth reading while you're twitchin for the UPS truck 

http://www.aikenamps.com/index.php/what-is-biasing

[dwinstonwood]

Thanks shooter. That was a clearly written read.

sluckey, I'm not on this forum to question or challenge your - or anyone else's - knowledge and instructions. Sorry if I've come across that way. I'm a rank beginner and you have the expertise and experience. And, you're generous enough to share it. I appreciate it.

I'm building the bias circuit with the 470 Ohm resistor, using the 50V AC bias tap on the PT secondary.

I don't understand it, but, like I said, I'm not questioning it. At this point it's all very confusing and I'm just trying to see how it works. I'm sure it will make a lot more sense when the amp is built and I'm actually measuring real-world voltages.

If I need at least -35V on the grids (pins 5) that's what I'll make happen. I have no idea why the schematic says -26V and I guess it doesn't matter; but, anyway, that's why I had chosen the 10K resistor. I'm not going to think about it anymore, other than that -35V seems safer for the tubes (when they're installed).

[What's so weird and confusing about this is this strange coincidence: with 375V on the plates the tubes would need 26ma of current flowing through them to be at 9.8 watt of dissipation. That number 26 could be what's mixiing me up.]

Enough for one day! I'm fried.

[sluckey]

I don't feel challenged at all. But if you don't understand what you're being told it's always good to ask questions. We all learn by doing so. You've received a lot of good info in this thread, but a lot of it is not directly aimed at building this 6G6, and a lot of it is way to deep for a beginner to firmly grasp. It's easy to become overwhelmed and begin to wonder how it all ties together. My approach is usually aimed at the specific topic at hand, like this bias tap/resistor value thing. You can read all the theory about load lines, different bias circuits, setting bias, etc., but in the end you will need to put a 470Ω resistor to the bias tap on that PT for this 6G3 project. Or you can use a 100K connected to the high voltage winding just like the original. Quick and direct to the point. Just git'er done! Just my style. No need to write pages and pages about it.

I'm building the bias circuit with the 470 Ohm resistor, using the 50V AC bias tap on the PT secondary. I don't understand it, but, like I said, I'm not questioning it.

I didn't just make up that 470Ω number. I copied Fender. Most all Fender amps from that era that used a 50V bias tap (and most did) used a 470Ω 1W. Take a look at this next generation Deluxe...

     https://el34world.com/charts/Schematics/files/Fender/Fender_deluxe_aa763.pdf

If I need at least -35V on the grids (pins 5) that's what I'll make happen. I have no idea why the schematic says -26V and I guess it doesn't matter; but, anyway, that's why I had chosen the 10K resistor. I'm not going to think about it anymore, other than that -35V seems safer for the tubes (when they're installed).

-35V is not the magic number. You may very well end up with -26V for your final bias setting, although I bet it will be more like -30 to -36. For this amp, I would like to make the bias range adjustable from -25 to about -40. Somewhere in that range will be your happy bias voltage. Make that range happen and you will be good to go when it comes time to actually adjust the bias for your desired plate dissipation. But for now, just set the pot for max negative voltage on pin 5 of the output tubes and you will not likely have any red plate surprises when you turn the power on.

[What's so weird and confusing about this is this strange coincidence: with 375V on the plates the tubes would need 26ma of current flowing through them to be at 9.8 watt of dissipation. That number 26 could be what's mixiing me up.]

Do not confuse -26V on the grid with 26mA. The voltages listed on Fender schematics are + 20%! And your 375V plate and 26mA current are just napkin doodles. Nice to talk about but just wait. It will be interesting to see what the REAL numbers are.

BTW, which PT did you order?

This has been a wordy post. My next post will be short and aimed at the specific question.   

[dwinstonwood]

Cool, thanks sluckey. That breakthrough moment of clarity when the puzzle pieces fit together and make sense is around the corner, I'm sure.

I bought the Hammond 290HAX:
https://www.hammfg.com/files/parts/pdf/290HAX.pdf

When I chose to build the 6G3 it was based on a deliberate and conscious decision not to build an AA763 Deluxe.

Most TX manufactures I looked into sell the same PT for the 6G3 and the AB763. The difference between 365V and 415V plate voltages seemed substantial enough to suggest different PT's were used in the original amps.

I'm just ready to start building. It's going to sound like an amp that I built. What that sound is...

[dwinstonwood]

A new amp arrived! This one has an orange theme.

The chassis is hardcore. It's more rugged than a truck bumper.

Still waiting on the TX's.

I ordered the parts before I knew I needed the 470 Ohm resistors, but two 1K 3W that measured low twisted together in parallel give me 490 Ohms 6W, so that worked out. A good omen?

This time I'm following the Hoffman sequence, wiring up first, and soldering the parts onto the board near the end. With the AC4 I did some things out of order and it made the build more difficult.

[sluckey]

This time I'm following the Hoffman sequence, wiring up first, and soldering the parts onto the board near the end.

That works fine for a turret board but is a terrible idea for an eyelet board. Which type board do you have?

[bmccowan]

Nice to see you getting started. I have two suggestions:

I ordered the parts before I knew I needed the 470 Ohm resistors, but two 1K 3W that measured low twisted together in parallel give me 490 Ohms 6W, so that worked out. A good omen?

Yes, that'll work. But you are building a nice new amp. And those resistors go right on tube sockets. So I suggest you order those resistors. You'll have them in plenty of time and you'll do it right the first time.
Second suggestion - listen to Sluckey about the eyelet board (and anything else) there is a reason for the acronym WWSD.

[dwinstonwood]

Thanks bmccowan. I also realized I overlooked ordering the 820 Ohm resistor in the LTPPI circuit. I ordered several 82K PI plate resistors, and thought I had everything. And, I need two 500pF caps for the pots. So, I'm going to have to do another order, anyway!

Believe me, I take sluckey's instructions very seriously. It goes back to the discussion above about not getting too focused. I stepped away for a couple of days in order to reboot my brain and clear its cache, too many processes running at once. I'm enjoying every post of this thread.

Yesterday, Uncle Doug posted a new video on a '63 6G3. I enjoyed watching that, and I got to actually see a real chassis layout and wiring. It's difficult comparing Fender's old "shorthand" layout drawing conventions/practices to whatever photos of real amps I can find on the web. DiyLC, they're not.

[dwinstonwood]

Well, this might be a major design mistake, but I'd like to install the filter cap board inside the chassis. It keeps it closer to the PT and power tubes with short wire runs (especially to ground).
I measured everything and drew it to scale, so there's plenty of room.
I also did my best to not cross power wires and signal wires on the same side of the board. Where they do cross on the board, they're on opposite sides.

There may be serious electrical issues with my layout, but it seems more straightforward with fewer power wires running back and forth to a doghouse. At least I understand where each connection is made.

[sluckey]

Nice looking layout. I didn't check the main board for errors but the control panel has many booboos. Better take a close look.

[dwinstonwood]

Thanks! Will do. These "digital dry runs" are really useful for me.

[shooter]

now all you have to do is increase your ready spares so you can quit spending beer money on shipping 

[dwinstonwood]

now all you have to do is increase your ready spares so you can quit spending beer money on shipping 

shooter, there's a serious danger with stocking parts. If all I need to build that "next amp" are the TX's and a chassis, what's to stop me from doing it?

[bmccowan]

what's to stop me from doing it?

So you are single

[shooter]

what's to stop me

 
here's my addicts list
eventually you get old and slow down 

[66Strat]

what's to stop me

 
here's my addicts list
eventually you get old and slow down 

What is a 369er? Is that similar to a 3:59er?

[dwinstonwood]

what's to stop me from doing it?

So you are single

there're +'s & -'s
Ability to fill the house with soldering flux smoke falling in the first column.

shooter, that's quite a list. I'll hopefully run out of money and/or space before friends need to stage an intervention.

[shooter]

that's quite a list

 
most were "college" lab projects, til I learn't enough to become half-ass-k

369'r got thumbs up from the guitar guys
3SE 6AU6 dissipating ~~ 9W in heat ~~ 4W audio 

[dwinstonwood]

Happy 4th everyone.

So, I'm trying to get started with installing components into the chassis. But, before I get too far along I want to be certain that I understand the grounding of the pots and input jacks - before I attach any wires.

I've spent a few hours looking at Fender layouts and photos online, but there are still some points of confusion. It seems like all the layouts I can find leave out one or two key bits of info that leaves me scratching my head.

Here's the best I can do on my own. If it's wrong, where am I screwing up? Thanks!

[sluckey]

Buss looks much better. I suggest these changes...

1. Break the buss so the INT and Speed pots are NOT on the preamp buss. INT and Speed will ground at the filter caps. Be sure to connect the filter caps and the PT HT center tap to chassis near the PT. Many people just use one of the PT bolts but I prefer a dedicated screw. OK to connect the filament CT to this same screw.

2. Put a jumper between the ground lugs of the input jacks as shown and also connect the jacks to the preamp buss as you already have it. Then connect the ground buss to chassis using a dedicated screw near the input jacks.

[dwinstonwood]

Awesome. Much appreciated sluckey!

I followed the original schematic for the speaker jacks and the tremolo jack. I've also uploaded the Mojotone wiring which is different. Neither indicate grounding the jacks to the chassis bolt. Maybe this is one of those things that's supposed to be so obvious that it's just not drawn? 

Should I run a ground wire from the speaker jack to the chassis bolt? I did this on the AC4 build.

Thanks again!

[sluckey]

I don't like to rely on the input jack body to provide the ground. Leo did. Hoffman does. But I've fixed several amps that had problems due to loose nuts on the input jacks. Providing a dedicated ground bolt totally eliminates this problem.

Same idea applies to the speaker jacks. Since your amp has a NFB loop from the speaker jack, the NFB circuit relies on the OT secondary to have a good ground. Run a wire from the speaker jacks to the power ground to provide a reliable ground connection for the NFB. Leo didn't. Hoffman did.

Bookmark this page...

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

[dwinstonwood]

Bookmark this page...

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

  That page, along with a few others have been on my bookmark toolbar for a couple of weeks now.
FWIW, I did see Hoffman's speaker jack ground wire, and had added it at one point. But, when I saw that Fender, Mojotone, and Weber's Lenora didn't specify it I guess I was compelled to removed it.

However, your explanation of why it adds an extra degree of reliability (a jack being a mechanical part that will work loose) makes it clear why to add it.

Thanks for the explanation. The things you're sharing are "good practice" concepts that I'll carry over into future builds.

On another note, the 290HAX PT is a good fit in the chassis. I had to elongate the mounting holes about 1/8" but it's a solid, rigid mount.

[bmccowan]

Take a look at your B+ nodes - I think you have a mistake feeding V2.

[dwinstonwood]

Take a look at your B+ nodes - I think you have a mistake feeding V2.

Thanks, bmccowan. I'm sure you're right. I was trying to follow the schematic attached. It looks like the tremolo plate of V2 is fed from the first node, before the 1K resistor, and the post-attenuation gain half of the tube is fed by the last node along with V1.
I'll trace the path on the layout and post that, too.

[dwinstonwood]

And, here's the way I interpreted the layout:

[bmccowan]

Thanks, bmccowan. I'm sure you're right.

Actually, I'm sure I'm wrong - and I've built one of these amps! I jumped too quickly - sorry.
Or as Emily Latella used to say; never mind.

[dwinstonwood]

Believe me, I had to stare at that zigzag path to the first plate of V2 for a long time before I finally saw it! In an earlier post I referred to "Fender's old "shorthand" layout drawing conventions/practices." That's why I needed to spend the time drawing a DiyLC version, so I could trace out every connection myself.

[dwinstonwood]

...Just be sure that you have at least -35V on pins 5 of the output tubes.

I just read this Valve Wizard page, and now my thick head finally has a grasp on this.
http://www.valvewizard.co.uk/bias.html
The mu factor for the 6V6GT is 10. So, if my screen voltage is somewhere between 350V and 375V I'll need at least -35V on the grid to reach cut-off.
Exactly what you told me sluckey! Hey, I wasn't doubting you, I'm just one of those obnoxious people who always has to asks, "why?"

I've started soldering up the control panel and have the tx's and most of hardware installed in the chassis.
I'll update with photos when I hit a wall.

[dwinstonwood]

Progress. Eyelet board is next.

[bmccowan]

Neat work! Have you soldered the ground buss to the back of the pots before? I know it works and is a common method, but I gave that up after seeing other methods. A couple of weeks ago I opened up an amp that I built 6 years ago and found that 3 of my solder joints on pot backs had come loose. I suppose I did not use enough heat, or didn't clean the backs of the pot well enough.