Scratch Build, -from the beginning...

i started a private conversation with andrewRneumann regarding some help with the https://www.vtadiy.com/loadline-calculators/loadline-calculator/
he was a tremendous help in walking me through the load line calculations for a "Vintage" build im planning on, using a pentode as the V1.
he suggested we take this conversation to the public forum so everyone can chime in, and/or gain some benefit from our discussions.
so here it is, copied, slightly edited to keep the jibber jabber to a minimum.

at the time of this post, we havent gotten too deep. but its a lot to post and read all in one shot.

==========================

Yesterday at 12:10 PM

Jewellworks

andrew, youve been an enormous help with the Pentode calculations, and im hoping for some more guidance, since your familiar with this calculator.
im looking at V2 now (6J5 =triode), but going back and forth from RobRobs page on how to draw load lines, and this calculator... which LOOKS simpler, im still lost.
RobRob says you get your operating point by combining the value of both the Plate load resistor, and the Cathode resistor. that seems simple enough, but when i use the numbers i had on my design, (copied from something Printer2 posted for a 6J5 in the V2 spot) my load line is VERY flat. which -according to RobRob, will get more gain, and less headroom.
the 6J5 calculator defaults to a Load Resistance of 12000. which may or may not mean anything, but it shows a much more vertical line, which indicated less gain, but more clean headroom. but if the 12000 DOES mean "something", then how do i know how much of that is the Cathode resistor, and how much is the Plate Load resistor? the "load" on the calculator is both, combined.
and then theres the actual operating point. im not entirely sure what im looking at, or where to set it.
in robrobs example on his website, the 0v grid line is the saturation point, and sets his operating point halfway between 0V grid curve, and the 250v at the bottom of the chart. in my trials (and errors) using 300v as my B+, i set my operating point at 200v, which is about halfway between 0v grid curve and cutoff at 300v, using 12000 as my Load Resistance. it gives me a reading of -5.92 Grid Voltage.
what does that even mean? we were trying to keep it under -1 with the Pentode, assumingly because were using grid leak biasing. but what does it mean for a triode with cathode biasing? the grid voltage in my tests above is at -5.92. thats where the operating point is set. i guess what i dont understand is... is that good? or bad? too high? to low? or just a "thing".
the whole point of this is to know what resistors to use, where, and ... well.. im lost. im sure there are calculations i need to make w Ohms Law, but i dont know what they are. i also dont know if my load lines are good, bad or otherwise.
im not wanting you to do it for me. i just dont understand how all this works and need someone to hold my hand and explain it. you were awesome with the Pentode. i hate to impose on you again for this, -but i guess i am...

Yesterday at 2:10 PM

andrewRneumann
Tele-Holic
Hi @Jewellworks,

I'm going to answer your question with another question! Sorry...

Have you designed the power amp yet? There is definitely an "order" of design that makes more sense than doing it from input to speaker. One thing depends on another and if you design parts in the wrong order, you will end up redesigning previous stages. So what I suggest is actually to start with you speaker and work backwards to the input.

What speaker do you plan to use?
What output transformer?
What voltage will the power tube run at?
What will the bias point of the power tube be?

Once you answer these questions, you can then determine a very important piece of information: What signal can the power amp accept before it reaches overdrive?

This is important because now you have a target. You can say, "I want my preamp to take my guitar signal, amplify it, and end up with something that will drive the power amp to full power." What I'm saying is, how can you design the preamp if the power amp isn't designed? You don't know what your target is. So I suggest taking a step back from the preamp, and learning everything you can about single-ended power amp design. Design your 6L6 and figure out what kind of signal it will take to take it to full drive. Then we will come back and look at you 6J5 and even redesign the 6SJ7 if we have to.

Guitar -> 6SJ7 -> Tone Stack -> Volume -> 6J5 -> 6L6
-17dBV +? dB -? dB -? dB +? dB Target dBV

Here is what I think you have for a signal chain. We are going to work in dBV instead of volts. A 200mV humbucker strumming all out is -17dBV. That's on the left. Target dBV to amplified by the power tube is on the right. How do we get from -17dBV to the target if we don't know what the target is?

Once we know the target, we can make estimates for how many dB your tone stack and volume control (it will depend on knob rotation) will cost you. Then we can go in and figure out what we need gain wise from the 6SJ7 and 6J5--these tubes add dB.

Read everything on Merlin's valvewizard website. Also ampbooks.com is very good. Aiken amps has some amazing information. I've probably used those three sites more than anything.
This is a learning process, it should be fun, but it takes work and a lot of reading and repetition before it starts making sense.

Andy Neumann

 

Yesterday at 3:02 PM


Jewellworks

you have the basic building blocks right for this design. my plan was to basically copy the Power section of a 5F1 Champ, but use a 6L6G instead, and bias as needed to keep the plate dissipation right at the limit, or just a tad over. i wont be touring stadiums with this lil thing. just bedroom plunking, and probably some recording. i have an 8 ohm OT that came from another SE amp with a 6L6GC. i thought id keep the power around 350v, but my PT can do close to 450v! so that could change, as needed... i dont have any idea what the dBv is.
ive read some of Merlins pages, but its confuses the hell out of me. it might as well be from Mars.

and i still dont know how to extrapolate the information i need from the vtadiy calculator

Yesterday at 3:50 PM


andrewRneumann
Tele-Holic

Ok, start with your power transformer. Figure out the actual filtered voltage you are going to have available to the 6L6. You say 450V is available, but was that peak AC voltage? If so, then the DC voltage after power supply filtering will be lower. If 450V was RMS voltage, then your DC may be higher than 450V. Either way, we need to know this. My experience with single-ended designs is a choke for the power supply is advisable, or else you will end up with 120Hz hum. Do some reading on Merlin's website about power supply design, smoothing and filtering!

This is to determine one thing: B+ available to the 6L6 plate and screen. Then if you have the OT specifications, you can draw a load line for the 6L6. Once you have that, we can talk about what the target dBV (output signal level) is for your preamp.

Steps Recap:
1. Actual filtered DC voltage available with your PT and rectifier.
2. Load line for 6L6 with your OT.
3. Figure out what signal level you need from your preamp.

This needs to be done before you worry about the nitty gritty details of the preamp. Show me the 6L6 load line!
You've got the 19W 6L6 or the 30W 6L6?

Yesterday at 4:27 PM


Jewellworks
Tele-Meister



https://frank.pocnet.net/sheets/127/6/6L6G.pdf
10.8 watt
mine is Sylvania branded, but this is easier to read than French...

power transformer
VOLTAGES: using a variac, i set the input to the primary at 120VAC

These PT tests were on the bench. Not in any circuit, no filter caps, nuthin

2ndary: HT: 862 VAC (!!)
HT-a to C/T: 432
HT-b to C/T: 429

6.3v Filament: 7.53
F-a to C/T: 3.77
F-b to C/T: 3.73

5v Heater: 5.77

in my 5F2a variant, i put a 22uf/450v filter cap at the 1st node of the rectifier, before the transformer to filter it. At first, i put it directly through the transformer and it hummed like mad. the 22uf cap fixed that. then i dropped the voltage w a 22k/2w and another 22uf filter cap to the screen. i just figured id do the same for this Vintage style amp.
otherwise... what can be done?
i dont have any OT specs. i just know that it came from an amp with an 8 ohm tap. i labeled everything before i took it out.



Yesterday at 9:27 PM


andrewRneumann
Tele-Holic

Please answer the following:

I just want to make sure you used a voltmeter that measures RMS voltage. If you measured the primary voltage at 120v then, it is most likely RMS. Can you confirm this?

Also, what did this PT come from? Any idea of how much current it can handle? Does it have any markings or anything on it to help identify?

What rectifier do you plan to use? Valve or solid state?

If you can afford to buy a choke, I would recommend planning one in your design, but we can probably make do without. We will get into sizing the resistors and caps later. Sizing the resistors is one of the last things to do.



Use your variac to determine the winding ratio of your OT. Apply 1v (measure it will your voltmeter--don't trust the variac label) to the secondary winding and then measure the volts across the primary winding. We will need this to draw a load line for the power tube.

Andy Neumann

 

@andrewRneumann -i have a Fluke 77 digital multi meter. i confirmed the voltage going in and coming out of the PT with the meter, and didnt even look at the reading on the Variac, as i wouldnt trust it. i DO trust my meter.
as for the OT, again, this was the original OT in a lil 6L6GC/SE amp that had a 70v, 25v and an 8 ohm tap on the 2ndary. im 100% certain it is what it says it is. i wont be able to do any measuring until later tonight, but the resistance across the primary is 270ohms. im positive it is what it says it is. i wont be using the 70v and 25v taps, and ill cap them so they dont short anything.
the PT came from an old Ampro projector using 3, 6V6GTs, the 6SJ7 and 6J5 in my Vintage build, and im using the 5Z4 rectifier that was in the Ampro. looks like from the 40's, 50's...
i really dont want to buy a choke for this amp. ive had good results using a 22uf/450v filter cap in my other 5/10w SE build. i want to use all leftover parts i have already on hand. especially the big pieces

 

i measured 10volts into the 2ndary of my OT, and i got 203v on the primary. did the math and thats a winding ratio of 20.36. squared, thats 414.52 impedance ratio. looking at the 6L6G data sheet, class A load resistance is 5000 (@ 270v). 5000/414.52 is 12.06. -hhhmmmm...
then i looked at the 6L6GC, which is what was originally tied to that particular OT. its load resistance (class A) is 4200. do the math, its 10.13.... hhhmmm again...
then i looked at the 6V6GT load resistance, Class A, =8500. math = 20.5! what?!?
then for giggles, i looked at 6L6G in AB mode. 3800. that comes out to 9.16.

none of these is 8 ohms, which is what the OT was originally used, and labeled as.

im going to have to find a Tube with a Load Resistance of around 3300. -or a different OT, with an impedance ratio of 625 to use with this 6L6G i want to use. -which sucks. especially since transformers dont come with that sort of information.

the only thing that might fudge the numbers is the volts listed for that load resistance is almost 100V lower than what i was planning on putting on the plate. so "maybe" i can still get away with using this OT?

or, maybe its "off" because of the other 70v and 25v taps? i cant see how, but... maybe?

 

or find an OT that has a Load resistance of around 3000 in Class A...

http://rover.ebay.com/rover/1/711-5...0001&campid=5338148343&icep_item=224143986360

https://frank.pocnet.net/sheets/168/7/7591.pdf

 

i thought once you divided the load resistance by the impedance ratio, that gave you the ohm reading for that 2ndary. 12.06 is an odd number with a 6L6G, and the amp it came from, with the 6L6GC comes to 10.13... -still not 8! or 7 something... so, did i miss a step?

ill read the merlin article you linked and get back to you

 

im a senior level AV tech, and i understand what the 70v and 25v taps are for. this OT came from a very small PA amp (only 10 watts) and those taps are for running whats called a "distributed" audio system. these kinds of systems are used in areas where there are multiple speakers, such as airports, shopping malls, theme park audio... you can string several speakers together, each one having its own transformer tapped at a certain wattage that it will use off the main feed. in this case, there were only 10 watts available, so my guess is they used 3 speakers at 3 watts each. or 8 or 9 speakers at 1 watt each, leaving a little headroom. its also how the telephone company phones work. 100v line with a 200 watt amplifier and your only running a telephone speaker tapped at 1/8 watt or less... you can string a LOT of phones together... the voltage on the line determines how long the run can be. in our industry, we use 70v all the time.

still deciphering Merlin's algebra... my brain is boiling

 

i "think" i got the first part of his Wizardry figured out...

"The source resistance is equal to the resistance of the secondary coil, plus the resistance of the primary coil when 'reflected' to the secondary:
Rs = Rsec + Rpri/(Vpri/Vsec)^2
Where:
Rsec = measured resistance of the secondary winding.
Rpri = measured resistance of the primary winding.
Vpri = RMS primary voltage (e.g., mains voltage).
Vsec = RMS secondary voltage measured with no load.

For a two-phase rectifier the voltages can be predicted in exactly the same way as above, except that we only need to consider half the transformer's secondary winding."

R2ndary= 217 (c/t to one side of the 2ndary. so half the winding)
Rprimary= 5.9 + 60? (=66)
"Also, if a valve rectifier is used, then its internal anode resistance also forms part of the total source resistance Rs, as shown. This is not constant, so just use a ball-park average. This can be obtained from the valve data sheet"
the 5Z4 data sheet (https://frank.pocnet.net/sheets/127/5/5Z4.pdf) says the "Total Effective Plate Impedance Per Plate" shows 30 ohms. since there are 2 plates, thats 60. but this data sheet (https://frank.pocnet.net/sheets/049/5/5Z4.pdf) shows a Minimum of 50, and a Maximum of 125 ohms, PER PLATE. sooooooooo... thats a huge difference.

anyway... 217 + 66 = 283
Vpri = 120 / V2ndary = 430 = .28
283/.28 = 1010
squared, thats 1021514, or 1.2M
RS = 1.2M

thats as far as i got, and i have no idea if this is even close.

the 2nd part, is complete gibberish

"The amplifier circuit (or whatever happens to be draining current from the reservoir capacitor) can be represented as a single load resistance, Rl. This can be estimated simply as:
Rl = Vdc / Iload
Where:
Vdc = DC supply voltage after rectification (i.e., the voltage across the reservoir capacitor).
Iload = average DC current demanded by the amplifier circuit.

If the amplifier runs in class A then its current demand does not change much with drive level, so as far as the power supply is concerned it looks like a constant resistance. A class AB amp, on the other hand, will demand less current at idle than at full drive, so its apparent resistance is not constant. Therefore, when estimating the load resistance, use whichever level of current you are more interested in. If you're really keen you could do two calculations, one for idle, one for full drive, so with the following method you could accurately predict the amount of voltage sag you will get under sustained overdrive...
Of course, to find the load resistance we need first to estimate what the DC supply voltage will be. If the circuit were ideal the capacitor would charge up to the peak AC voltage, or Vrms*sqrt2. This should be close enough as most amplifiers will adjust their idle current to suit the supply voltage anyway, so they are more-or-less resistive."

-Valvewizard

every single word of this is gibberish to my mind. i have no clue what im supposed to be doing.

HELP!

 

he doesnt say how to get the Current demand. only that its constant in a Class A circuit. he says the cap should charge up to the peak AC voltage, which should be 430, as thats half the voltage measured at the C/T. he also gives a gibberish algebra formula, but acts as it its the same thing as the half voltage. so why bother with the math? but he never says how to get the Iload or current draw. the rest of the page is gobbledygook

 

I was mistaken in that post. It's 217

 

https://www.tdpri.com/threads/power-transformer-readings.1049579/