TWEED TWI-LUXE, FINISHED!

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
Well folks,
Before digging in to building a JTM45, I need to get this project done and off of my bench. If it turns out like I hope, it may negate (for a while, at least :rolleyes:) the need for said JTM45! :naughty: Its gonna be a low powered Tweed 5E8-A Twin preamp into a cathode bias Tweed 5E3 Deluxe power section. I do have a few questions to answer though, concerning the input stages. Here's a "married" schematic:

Twi-Luxe-5E3-A-Schematic.jpg

Notice that at the inputs, there are no grid stop resistors and that when using the #1 input of either channel, we get an effective grid "leak" value of 500kΩ and that input parallels to both triodes of that tube/valve. My plan is to also parallel to the input #1 of the other channel. My questions concern what value I should use for a common grid leak? I will leave open the ability to add grid stops later, if it seems appropriate.

My thought process here is that by the time the signal gets to the grid of V3A of the cathode follower pair, the signal will be quite strong, but with lots of harmonic and tonal character. I'm guessing the paralleling will not produce an overly high amount of gain, but instead a very strong and wide ranging signal? Getting my voltages right will likely be a challenge! The transformers are Classictone 40-18085 PT,
http://www.classictone.net/40-18085.pdf
MPS OT20PP-VC87, 8K/6K6
And Mojo777 choke
https://www.mojotone.com/amp-parts/Filter-Chokes/Filter-Choke-for-6L6-Amps

I'm not sure if I'm going to actually use the choke at all and if I do, whether or not I should put it before the OT/CT or between that node and the screen node?

For reference, here's a link to a schematic for the complete, original 5E8-A:

https://cdn.shopify.com/s/files/1/2325/2729/files/Twin-5E8-A-Schematic.pdf?5926397210715640943

Notice how and where they use the choke and also that there is no separate node for the screens and no sceen resistors. I'm concerned that it's going to be difficult to keep the B+ up to around 370-ish and also keep the voltages of the earlier stages up at the listed values. I may even need to create an additional node for V1 & V2?

While I've got most of the hard components mounted, I want to get all my planning done before heating up my iron! Obviously, I may need to tweak and twiddle the resistors in the dropping string a bit!

Any and all thoughts and comments will certainly be welcome here.

Thanks Folks,
Gene
 
Last edited:

thetragichero

Well-Known Member
Joined
Jun 6, 2010
Messages
3,426
Reaction score
1,375
Location
upstate ny
definitely add screen resistors to the power tubes. 470r seems to be the standard value used on later models. no reason to leave them out
boy that's an inefficient use of tubes on the preamp! if expect that first stage to be on a pa rather than a guitar amp: four inputs exactly the same besides a small bright cap on two of em
if it were my build, I'd use the later fender high low inputs (also answers your preamp grid resistor and grid stopper question. if you're still going to share the cathode use 820r), freeing up an entire tube for.... reverb? tremolo (i happen to like the one from the vibrochamp)? extra gain stage (since this is using 12ay7s instead of 12ax7s) and a long tail pair phase inverter? i guess that would make it less tweedy

but yeah as it is I'd consider it a waste of a tube and jacks with the inputs like the schematic
 

_Steve

Well-Known Member
Joined
Apr 29, 2020
Messages
564
Reaction score
477
Location
Los Angeles
Ive always wanted to try choke-fed power tubes - will be interested to see how it turns out! Make sure your choke can handle all the current.

Ive built quite a few 6V6 amps out of Hammond organs and although im nowhere near an expert i'd suggest these:

1. Go back to fixed bias, or at least put a toggle switch between fixed and cathode/doggie-doodoo biased. It'll tighten up those 6V6s and give you a surprisingly snarly, punchy sound. Copy the 6G3 schematic for the bias circuit values.
2. Agree with @thetragichero above, I can't see any advantages to that input section unless you were plugging multiple instruments in.
3. Stick with the Cathodyne phase inverter. Theres no benefit to an LTP in a 6V6 amp. I've compared both in identical 6V6 circuits. If you are making this really high gain, then there's a trick on Merlin's page of adding a large grid stopper (~500k) which eliminates all the weirdness associated with Cathodyne distortion.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
definitely add screen resistors to the power tubes. 470r seems to be the standard value used on later models. no reason to leave them out
boy that's an inefficient use of tubes on the preamp! if expect that first stage to be on a pa rather than a guitar amp: four inputs exactly the same besides a small bright cap on two of em
if it were my build, I'd use the later fender high low inputs (also answers your preamp grid resistor and grid stopper question. if you're still going to share the cathode use 820r), freeing up an entire tube for.... reverb? tremolo (i happen to like the one from the vibrochamp)? extra gain stage (since this is using 12ay7s instead of 12ax7s) and a long tail pair phase inverter? i guess that would make it less tweedy

but yeah as it is I'd consider it a waste of a tube and jacks with the inputs like the schematic

Yeah, I get that whole "wasted efficiency" perspective. On the other hand, there appears to be a lot of love for the character provided by the parallel use of the two triodes in either of the channels. I'm simply looking at adding to that the character that gets provided by running two channels in parallel on separate volme controls, as in a Marshall with the channels jumped, but with two parallel triodes per channel! I will likely separate the cathodes of the two pairs of triodes. I know its kind of a goofy direction, but I've not yet found anyone who's tried what I'm planning! I'm also operating on the assumption that while related gain and actual signal strength are not exactly the same thing? I'm guessing that while the signal at the outputs of the volume pots will be fairly "strong" but not a lot of gaiin, per se.

Thanx 4 Lookin'
Gene
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
Please forgive me if I have committed a "faux pas" by bringing up a discussion of a (Heaven Forbid) Fender amp model to this section of the forum. :eek: I figured a Tweed Twin certainly qualifies as a "Classic" and I'm "Building" it so.......? :shrug: Not to mention, this is the section where all the cool kids, who know the differences between Sh!t & Shinola tend to hang out! :dude:

I'm currently endeavoring to get someone who actually owns, specifically a 5E8-A Tweed Twin to do a few operational tests for me to either confirm or dispel my suspicions of how it should work. After perusing all the different input and parallel triode configurations that Leo and his crew experimented with, I'm suspecting that they either A) missed a really cool feature, and/or B) determined that feature to be of no practical use, at that time! Today is a different time! I'd really like a few of my questions answered before committing to a full scale build and heating up the iron!

In this world of modern amps, many seem to have tunnel vision towards GAIN, but that's not all there is to having a strong, rich signal. You see, what I'm looking for is clear (not necessarily "clean"), unadulterated, complex harmonic feedback, WITHOUT brain searing gain and distortion or paint peeling volume!

FWIW, Matchless and a few other high priced boutique builders do indeed utilize parallel triodes to great effect. I won't go into all the details of the what, why and how, until I confirm validity. Then and if, I'll shout it all from the roof tops!

Thanks For Tolerating Me!
Gene
 
Last edited:

thetragichero

Well-Known Member
Joined
Jun 6, 2010
Messages
3,426
Reaction score
1,375
Location
upstate ny
nothing wrong with lower gain(distortion), fender-type stuff. and nothing wrong with parallel triodes (I've got a build that uses em) but four is kind of a waste (not to mention having three y-cables sticking out the front, and having to remember to pack them)
using just two triodes and the later fender high/low jacks all you'd need would be a jumper cable. more unnecessary tubes means more stuff to go wrong and more tubes to replace. KISS
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
nothing wrong with lower gain(distortion), fender-type stuff. and nothing wrong with parallel triodes (I've got a build that uses em) but four is kind of a waste (not to mention having three y-cables sticking out the front, and having to remember to pack them)
using just two triodes and the later fender high/low jacks all you'd need would be a jumper cable. more unnecessary tubes means more stuff to go wrong and more tubes to replace. KISS

On the stock unit, only one "y-cable" is required. Look closely at how that jack switching works. If I can confirm proof of concept, I will wire four input jacks to accommodate a simple single jumper., while still leaving out any grid stoppers. I'm guessing I'll need to play around with the values of those input resistors, which effectively act as grid "leak" resistors, in this case. My thought is that I'm gonna lose 3db of signal at the input, but get that much and maybe a little bit more by the time it gets to the grid of V3A. At that pint the signal should be a bit stronger/higher voltage than with just one pair of triodes and yet increase the dynamic range and tonal complexity a bit. Then again, maybe I'm just full of POO? :D Either way, I want to try it, and/or at least get someone who has access to one of these very unique amps to try it and report back to me the results!

Thanks For Your Input! :)
Gene
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
Well I've had a couple folks at some other sites do some testing for me, on their bone stock 5E8-A Twins. They Y-Jacked the two #1 inputs and were amazed at the response and generally confirmed my suspicions and concept. Once I work out a few little details, I'll be sniffin' sodder fumes!

And oh yeah, I misspoke about no separate node for the power tube screens! Duh! :nuts:

My plan is to first build and fire it up with no choke and no presence, but instead a standard Fender 25/25 cathode bypass cap on V4A. Next step, once all is stable, will be to split the cathodes of the two triode pairs of V1 & V2. Then, I'll work out the best presence arrangement (it'll be off the 4 ohm OT tap), and then add the choke, either where drawn, or just past the OT/CT. I'm going to try both locations, for reference. I'm still torn as to where the passive effects loop and master volume should be. On a 5E3, the loop works very well (for my uses) just before what is V4A on this amp, but is actually V2A on a 5E3. I'm toying with putting it between V4A & V4B, and/or at least putting the master volume there? I've included a few extra spaces and turrets on the board to accommodate these tweaks & twiddles! If it gets too ugly after moving things around too much, I can aways just make up a new board tailored to confirmed specifics!

Please remember that I'm not really a tech, but instead simply a guy with just enough knowledge to be dangerous and a pain in the @$$. My work is always solid, even though the cosmetics leave a bit to be desired. I have been repairing various bits of solid state sound gear for years, but doing anything with tubes, other than constantly playing them, is fairly new to me. This will be my 5th build and the first one all sourced and planned from scratch, by me. This is why I'm bothering you good folks with tons of questions. I'm not attempting to come up with some brilliant, "new," Earth shattering amp design, but rather an unusual combination of features, tones and responses that years have playing have led me to feel is best for me! Some pics of builds #3 & #4:

5E3


IMG_1122.JPG IMG_1121.JPG


THE BALLZZ - 10 watt 6AQ5


IMG_2059.JPG IMG_2065.JPG IMG_2064.JPG
FWIW, that 6AQ5 THE BALLZZ was the chassis and iron donor for this current build!
I'm Pretty Stoked!
Gene
 

thetragichero

Well-Known Member
Joined
Jun 6, 2010
Messages
3,426
Reaction score
1,375
Location
upstate ny
clc filter before the plates make sure it's rated for at least the full dc current that will drawn by the amp. will significantly reduce ripple going to tho the entire amp. is that desirable? this leads me to believe maybe not: https://www.ampbooks.com/mobile/classic-circuits/class-AB-ripple/ . might be worth hearing with your own ears
the big benefit to having the choke between the plate supplies is that it's 'not a resistor'. great filtering without the big voltage drop. also allows for a much smaller choke due to significantly less current draw
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
clc filter before the plates make sure it's rated for at least the full dc current that will drawn by the amp. will significantly reduce ripple going to tho the entire amp. is that desirable? this leads me to believe maybe not: https://www.ampbooks.com/mobile/classic-circuits/class-AB-ripple/ . might be worth hearing with your own ears
the big benefit to having the choke between the plate supplies is that it's 'not a resistor'. great filtering without the big voltage drop. also allows for a much smaller choke due to significantly less current draw

That's exactly why I'm going to save the choke for last and play around with a couple different scenarios "AFTER" all the rest of the meat & potatoes is well worked out! And thanks for the link, I'll check it out!

Still working out my input jack arrangement and input resistor values. I know what I want it to do, but a little confused on how to get there! I want to use two jacks. One jack (top) accesses both pairs of triodes (channels) with a combined input impedance of 250 kΩ. The other jack (bottom) will access only one triode pair with a 500 kΩ input impedance, while the top jack now accesses only the other pair of triodes, also with a 500 kΩ input impedance. Just not sure yet how best to wire that up? I might end up buping those impedances up to 500 k and 1 meg, respectively. I'll add grid stops, if it seems necessary. We'll see!

I know this is an unusual topography and such, but I really think it will work well for what I want to accomplish! And as many have come to realize, what is electrically correct isn't always what sounds the best! Us guitar Screechers & Squawkers can be a strange bunch!

Thanks Again For Your Help!
Gene
 

adew1

Active Member
Joined
Jul 18, 2020
Messages
259
Reaction score
190
Location
Switzerland
Watching this with interest. I like to see experimentation and people trying new things.
 

thetragichero

Well-Known Member
Joined
Jun 6, 2010
Messages
3,426
Reaction score
1,375
Location
upstate ny
grid stoppers on input? always necessary. I'd definitely not go as low as 250k input impedance as the guitar volume/tone controls also play a role in the impedance (they're in parallel with your amp input). too low and your signal will be loaded down too much
but yeah, use grid stoppers before every stage, even if they're small (fender-style amps I'll use 10k on stages after the first)
 

Travis398

Well-Known Member
Joined
Dec 7, 2011
Messages
1,318
Reaction score
1,902
Location
USA
I was thinking the same thing. But I'm not entirely sure what Gene is thinking, so I say go for it experiment and see.


But if you lower your input impedance too much your signal just gets grounded out, and won't be very loud.
The grid stoppers chop off the high frequency signals and allow the guitar signal to go through.
Eliminating unused high frequencies helps stop parasitic oscillation and RF interference.

Not discouraging you from trying this though. I built one of these old circuits myself, and liked it.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
grid stoppers on input? always necessary. I'd definitely not go as low as 250k input impedance as the guitar volume/tone controls also play a role in the impedance (they're in parallel with your amp input). too low and your signal will be loaded down too much
but yeah, use grid stoppers before every stage, even if they're small (fender-style amps I'll use 10k on stages after the first)

I was thinking the same thing. But I'm not entirely sure what Gene is thinking, so I say go for it experiment and see.


But if you lower your input impedance too much your signal just gets grounded out, and won't be very loud.
The grid stoppers chop off the high frequency signals and allow the guitar signal to go through.
Eliminating unused high frequencies helps stop parasitic oscillation and RF interference.

Not discouraging you from trying this though. I built one of these old circuits myself, and liked it.

Well guys, if you'll notice, my first intention is to build the preamp and it's inputs to duplicate how a stock 5E8-A operates as two individual channels plugged into the High inputs and the way the same operates when Y-jacking those two channels. IE: 500k and 250k respectively, with no grid stops (as per original schematic/design) and I will be leaving the option, turrets and real estate for experimentation and modificationas deemed appropriate. I really appreciate all of the comments that help give me some perspective and ideas for what to try once it is all up and running! All mods will be done individually to ascertain what each one accomplishes and then decide which of those may be beneficial to combine for further analysis! The really great thing that I'm looking forward to here, is the absolute requirement to spend endless hours playing all of my favorite guitars to confirm and/or dispel my various assumptions!

I will bother you kind folks by keeping you posted as to my results and to likely ask questions about how to best accomplish some of those experiments and modifications!

Thanx 2 All!
Gene
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
@_Steve
First, thanks for your offer to help! :cheers:

Next, I seem to be over the current capabilities of the Power transformer on hand, at least according to:

https://thesubjectmatter.com/calcptcurrent.html

I do however question it's accuracy, as when I plug in other numbers from other amps I have on hand for testing the voltage is considerably higher than my known and tested readings? Also, the choke I have on hand is 3.5H/100ma/100Ω. Others have helped me calculate my current draw, as per tube choice in my updated schematic below, and we've come out to about 90ma. The ClassicTone 40-18085 specs at 630v/100ma:

http://www.classictone.net/40-18085.pdf

I've got no problem getting different iron, but how to choose the right one that won't be too stiff is a bit beyond my expertise. I've been led to believe that part of the magic of many Tweed designs is having just barely enough current available to get the job done without melting down! A 5E3 is one example. And while I want to minimize hum and ripple, to a point, reading this short article cautions me to not want to completely remove ripple in the power supply:

https://www.ampbooks.com/mobile/classic-circuits/class-AB-ripple/

I also want to be careful about "stiffening" all the response too much in the filtering, as I kinda like squishy, within reason.

Understand that I'm simply a guitarist who knows how to use a soldering iron and follow instructions, along with understanding safety precautions and how to take good measurements. I have a scope, good Fluke DMM, signal generator and dummy load It's late enough in my life that I don't really have time to study, experiment and learn a whole new trade of designing and building amps. On the other hand, I've played through enough 100s of amps to have noticed what I like and don't like, as well as being able to generally discern whether certain things are happening in the preamp or power section. Even though what I think I want isn't very complicated, nobody currently makes what I want! I just want to get to play through the amp I've come to believe is best for me! This is what has led me to this point!

To test my hypothesis, I've installed a passive effects loop (just two jacks) just prior to V2A in my otherwise stock 5E3. This has allowed me to drive various other amp's power sections with the 5E3 pramp as well as drive my 5E3 power section with many other preamps. It has confirmed that I love the way the 5E3 power section operates, even with all its warts, crooked teeth, crossed eyes and stray hairs. It has also confirmed that while the 5E3 preamp is glorious in many respects, it lacks some of the control I'd like to have over what it is sending down the line to the rest of the amp. It's not just about overdrive and/or distortion, but instead giving control "on board" the amp to shape the tonal color and character of that overdrive as it proceeds through the circuit. I've determined that these two dual parallel triode channels, also run in parallel with each other (and likely splitting the cathode pairs of each channel apart from each other), through somewhat of a Bandaxall style tone stack will probably get me where I want to go! And then on to my gloriously "farting" 5E3 power section, adding its own special color and character to the whole schmear!

And while I totally get what @Travis398 said about getting the input impedance too low decreasing the signal, that's what the four triodes in parallel will likely make up for. My plan for testing (once the amp gets fired up) is to put two 1meg pots in place of the input resistors to find the "sweet spot" measure the pots and install fixed resistors.

And then on the plus side, if it turns out that my "hunch" is full of poo, it's a walk in the park to remove a tube/socket and turn it easily into a cathode bias 6V6 5E7 Bandmaster, 5E5-A Pro, 5E4 or 5F4 Super!

Where I need the most help is getting my power supply in order, with enough wiggle room to experiment a bit! The power transformer I have is right on the edge of doing the job, but I'm just not sure! What's your take on my proposed current draw? FWIW, most of the voltages are in the ballpark of what I'm shooting for. The OT/CT B+ could go up a few and I may need to trim down the B+3/PI node a tiny bit.

Thanks Again,
Gene
Twi-Luxe-5E3-A-Schematic copy.jpg
 
Last edited:

_Steve

Well-Known Member
Joined
Apr 29, 2020
Messages
564
Reaction score
477
Location
Los Angeles
@_Steve
First, thanks for your offer to help! :cheers:

Next, I seem to be over the current capabilities of the Power transformer on hand, at least according to:

https://thesubjectmatter.com/calcptcurrent.html

I do however question it's accuracy, as when I plug in other numbers from other amps I have on hand for testing the voltage is considerably higher than my known and tested readings? Also, the choke I have on hand is 3.5H/100ma/100Ω. Others have helped me calculate my current draw, as per tube choice in my updated schematic below, and we've come out to about 90ma. The ClassicTone 40-18085 specs at 630v/100ma:

http://www.classictone.net/40-18085.pdf

I've got no problem getting different iron, but how to choose the right one that won't be too stiff is a bit beyond my expertise. I've been led to believe that part of the magic of many Tweed designs is having just barely enough current available to get the job done without melting down! A 5E3 is one example. And while I want to minimize hum and ripple, to a point, reading this short article cautions me to not want to completely remove ripple in the power supply:

https://www.ampbooks.com/mobile/classic-circuits/class-AB-ripple/

I also want to be careful about "stiffening" all the response too much in the filtering, as I kinda like squishy, within reason.

Understand that I'm simply a guitarist who knows how to use a soldering iron and follow instructions, along with understanding safety precautions and how to take good measurements. I have a scope, good Fluke DMM, signal generator and dummy load It's late enough in my life that I don't really have time to study, experiment and learn a whole new trade of designing and building amps. On the other hand, I've played through enough 100s of amps to have noticed what I like and don't like, as well as being able to generally discern whether certain things are happening in the preamp or power section. Even though what I think I want isn't very complicated, nobody currently makes what I want! I just want to get to play through the amp I've come to believe is best for me! This is what has led me to this point!

To test my hypothesis, I've installed a passive effects loop (just two jacks) just prior to V2A in my otherwise stock 5E3. This has allowed me to drive various other amp's power sections with the 5E3 pramp as well as drive my 5E3 power section with many other preamps. It has confirmed that I love the way the 5E3 power section operates, even with all its warts, crooked teeth, crossed eyes and stray hairs. It has also confirmed that while the 5E3 preamp is glorious in many respects, it lacks some of the control I'd like to have over what it is sending down the line to the rest of the amp. It's not just about overdrive and/or distortion, but instead giving control "on board" the amp to shape the tonal color and character of that overdrive as it proceeds through the circuit. I've determined that these two dual parallel triode channels, also run in parallel with each other (and likely splitting the cathode pairs of each channel apart from each other), through somewhat of a Bandaxall style tone stack will probably get me where I want to go! And then on to my gloriously "farting" 5E3 power section, adding its own special color and character to the whole schmear!

And while I totally get what @Travis398 said about getting the input impedance too low decreasing the signal, that's what the four triodes in parallel will likely make up for. My plan for testing (once the amp gets fired up) is to put two 1meg pots in place of the input resistors to find the "sweet spot" measure the pots and install fixed resistors.

And then on the plus side, if it turns out that my "hunch" is full of poo, it's a walk in the park to remove a tube/socket and turn it easily into a cathode bias 6V6 5E7 Bandmaster, 5E5-A Pro, 5E4 or 5F4 Super!

Where I need the most help is getting my power supply in order, with enough wiggle room to experiment a bit! The power transformer I have is right on the edge of doing the job, but I'm just not sure! What's your take on my proposed current draw? FWIW, most of the voltages are in the ballpark of what I'm shooting for. The OT/CT B+ could go up a few and I may need to trim down the B+3/PI node a tiny bit.

Thanks Again,
Gene
View attachment 86594

Hey Gene,

Firstly your PT is marginal. It's got enough current for the B+, but it is 10% underrated for your filaments because you have the extra preamp tube. When I ran the calculator I got 1.89A for 3 tubes and 2.22A for 4.

You should be OK with it but I don't want to give you advice then you blow your PT. Usually your filament voltages will come out a bit higher than 6.3V with US wall voltages being what they are so as the voltage goes up the current draw will go down a bit. You can go ahead with this design and monitor the temperature of the PT. In your design pulling out one of the first stage tubes will bring you back in spec if you start to have concerns.

For the power supply your approach should be to copy the design of the power amp (5E3 in this case right?). You need to fire it up, take voltage readings, then tweak. In your case it's most likely only going to be the last 8uF cap and its dropping resistor that may need to change. I'm guessing you may want to up that cap to 16uF. For the dropping resistor once it's up and running you can measure the voltage drop across it to work out what the actual current going through it is - then using V=IR (or R=DesiredVoltage/I) you can work out what resistor will give you the desired voltage drop. It's unlikely that the upstream nodes will need tweaking, but if they do its the same easy process.

One thing to bear in mind is that, unlike power tubes, 12A*7 tubes are super resilient when it comes to plate voltages. You can throw pretty much anything at them and they just work. Most people dont stress too much about their exact voltages, but this is probably a good exercise for you to go through to experiment and draw your own conclusions.

Just one more note. Don't stress too much about the PT "sound". The rectifier and the RC values are far more impactful. The 5Y3 is a very saggy rectifier.

For reference here is the output from the calculator when I ran it. It looks correct to me:

Selected transformer voltage: 315-0-315

Calculated voltage at first capacitor (B+): 393.75V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 2.22A

Calculated current: 90.21mA at 12920R calculated load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode​

Hope this helps! Let me know if i missed anything.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,533
Reaction score
3,988
Location
Las Vegas, NV
Thanks for that @_Steve
It looks like when you ran the calculator, you may not have entered the 8000R output transformer primary. I get the same and that might explain the 90--ish others have suggested. When putting in that 8K OT, I get this:

Selected transformer voltage: 315-0-315

Calculated voltage at first capacitor (B+): 393.75V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 2.22A

Calculated current: 126.21mA at 8000R load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode.
Also, the 6.3v filament tap of my PT is actually rated at 2.25ma, so that's fine. It doesn't matter though, as the Calculated current appers to be more than 25% to high.

Of course the really funny thing is that when I plug in all the info for my 5E3 Deluxe, also 8KΩ OT, I get this:

Selected transformer voltage: 355-0-355

Calculated voltage at first capacitor (B+): 443.75V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 1.56A

Calculated current: 124.12mA at 8000R load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode​

That "Calculated voltage" is way outa whack, as measured voltage in my 5E3, at the first filter cap is about 377V. Then when I leave nothing in the "Rload" OT field, the reading defaults to a 16410R load at 68.63mA. Notice that when you ran it without entering an "Rload" it calculated a load of 12920R. Note also that whenever an Rload value is specified, you get the same "Calculated current", no matter what power tube you choose. This is what makes me question the accuracy and/or validity of that calculator! It is also the source of much of my confusion? It really seems as though the OT primary impedance is an important part of the equation/calculation? Or is it?

Thanks,
Gene
 

_Steve

Well-Known Member
Joined
Apr 29, 2020
Messages
564
Reaction score
477
Location
Los Angeles
Thanks for that @_Steve
It looks like when you ran the calculator, you may not have entered the 8000R output transformer primary. I get the same and that might explain the 90--ish others have suggested. When putting in that 8K OT, I get this:

Selected transformer voltage: 315-0-315

Calculated voltage at first capacitor (B+): 393.75V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 2.22A

Calculated current: 126.21mA at 8000R load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode.
Also, the 6.3v filament tap of my PT is actually rated at 2.25ma, so that's fine. It doesn't matter though, as the Calculated current appers to be more than 25% to high.

Of course the really funny thing is that when I plug in all the info for my 5E3 Deluxe, also 8KΩ OT, I get this:

Selected transformer voltage: 355-0-355

Calculated voltage at first capacitor (B+): 443.75V. Subtract 6V, if using choke.

Calculated filament current (typically the 6.3v secondary): 1.56A

Calculated current: 124.12mA at 8000R load (10% plus factored in). Preamp valves current draw is estimated at typical 12AX7 max dissipation of 1.2W at 330V, i.e. 3.6mA per triode​

That "Calculated voltage" is way outa whack, as measured voltage in my 5E3, at the first filter cap is about 377V. Then when I leave nothing in the "Rload" OT field, the reading defaults to a 16410R load at 68.63mA. Notice that when you ran it without entering an "Rload" it calculated a load of 12920R. Note also that whenever an Rload value is specified, you get the same "Calculated current", no matter what power tube you choose. This is what makes me question the accuracy and/or validity of that calculator! It is also the source of much of my confusion? It really seems as though the OT primary impedance is an important part of the equation/calculation? Or is it?

Thanks,
Gene

I dont know why I thought your filament rating was 2A.

Forget the OT impedance. I cant even see how that's useful without other info such as your bias setup, which in your case is cathode biased so you're power draw is low. Lots of cathode-biased 6V6 push-pulls run from 100mA PTs. The numbers I got from the calculator seem exactly inline with what i'd expect and also validated by the PT datasheet (see the expected B+ voltages).

Honestly I think you're over thinking this :) Just start with the power supply design from a 5E3 (correct me if thats not the power amp design you're planning). As long as the power supply is good for the power amp section you are 95% of the way there. You'll only be tweaking the preamp voltages (and a cap), and perhaps an initial drop if the raw B+ is too high, which is quite common. We can cross that bridge when we get to it.

How are you building this? In a Fender style chassis with the dog-house thingee for the filter caps?
 
Last edited:

_Steve

Well-Known Member
Joined
Apr 29, 2020
Messages
564
Reaction score
477
Location
Los Angeles
Ah i've just re-read your initial post. So you are in fact using the power supply design from the 5E8-A but with 6V6s. This does mean you'll need to think a bit more about the design but its still not too hard.

Firstly, are you sure this is the way you want to go? Which feel (dynamics/responsiveness) are you aiming for - the 5E3 or the 5E8-8? Getting the 5E3 feel will be easy, the 5E8-A more difficult and only an approximation.
 


Top