I want to use elevated heaters in a Marshall amp. Help.

[cluster]

This is the PT I'm using...

JTM45-RI-P (Dual Primary - current version).pdf

[Helmholtz]

That's why we need a schematic - at least of the power supply.

Generally I'd always try a humdinger (heater balance pot) first before considering elevation or even DC heating.
If necessary humdinger and heater elevation can easily be combined.

[Enzo]

Well obviously if the amp doesn't even HAVE a bias winding, then I can't repurpose it. But I don't think we have established what transformer he is using on his build. And when a bias winding is present, I'll still prefer drawing my elevation from something like a 60v supply rather than a 400v supply.

[Helmholtz]

Probably late in the game, but I never liked dropping from hundreds of volts to get a few. Were it mine, I'd be inclined to add a second diode facing the other way at the bias tap and make a small low voltage positive supply for the purpose. Bonus: no way for some failure to put B+ in your heater circuit.

With amps like the JTM45, where the bias is derived from the HV winding (e.g. 350VAC), this method doesn't look much safer..

[cluster]

I've attached the elevated heater section in that Marshall schematic link that someone posted. Seems like it attached to the preamp can cap for reference. Any comments?

[pdf64]

….Using the BIAS tap as a positive reference point…
Someone mentioned using the bias tap as a reference tap because of its lower voltage. I looked at lord valve's schematic, so I get the idea. …

I don’t, I’m totally confused by what you’re getting at above.
Are you thinking of a negative heater ‘elevation’? That would be opposite to the usual goal.

I see that Friedman used the 16/16uf can cap as reference point for his elevated heaters in his Dirty Shirley build (see earlier photo). …Is using the can-cap as reference point a bad idea and why? What would be the best place to insert the elevated voltage divider? …

A can cap is just an unreferrenced pair of components, it doesn’t pinpoint a particular circuit node.

The screen grid HT node has been suggested earlier as a suitable feed for the heater elevation potential divider.

[cluster]

I was away for a couple of days, so didn't see the generous responses. Thanks so much .


The Build:
The build i'm working on is based on a Marshall JTM45 / 1987X / Frankenstein Hybrid. JTM45 power section (sort of) with a 1987x preamp (and tone stack) with tweaks. I also want to experiment with cascading the preamp for more gain. Also, I'll probably want to experiment with different V1 cathode values and maybe plate resistor values. Will bump the the power filtering to 50/50uf across the power rail to stiffen the bass response.

The Elevated Heaters:
As part of my eduction I want to experiment with elevated heats and maybe even DC heaters. I don't know much about setting up elevated (or DC) heaters in a Marshall type power section.

Using the BIAS tap as a positive reference point:
Someone mentioned using the bias tap as a reference tap because of its lower voltage. I looked at lord valve's schematic, so I get the idea. My question: is there an advantage to using the bias tap vs the HT as a reference point or does it even matter?

Once I decide which tap to use (low voltage vs high voltage) I'll provide a schematic.

I see that Friedman used the 16/16uf can cap as reference point for his elevated heaters in his Dirty Shirley build (see earlier photo). From what I gather Friedman is not a very well respected amp builder among the amp builder illuminati and made quite a few mistakes on his early builds. My question: Is using the can-cap as reference point a bad idea and why? What would be the best place to insert the elevated voltage divider?

Thanks again

[Gregg]

I always use transformers with dedicated bias secindary and never use it for anything else unlike for example mesa where you can see all kind of voltages derived from it. I usually put fuses on all secondaries except on the bias one.
In theory all kinds of failures are probable in an amp and it's not possible to implement protection for all of them.

[Enzo]

Probably late in the game, but I never liked dropping from hundreds of volts to get a few. Were it mine, I'd be inclined to add a second diode facing the other way at the bias tap and make a small low voltage positive supply for the purpose. Bonus: no way for some failure to put B+ in your heater circuit.

[Chuck H]

The OP never indicated that the jvm410 was even similar to the amp in question. It was strato56 in post #3 that posted the schematic as an example of how Marshall did it with that model. Although there may be a good solution to the OP's problem by perusing the schematic, there may not. We need more information about the actual amp in question before a solution can be arrived at.

[tedmich]

marshall_jvm410.pdf Schematic (I couldn't find in thread) you have to click "insert" after uploading apparently : (

[Gregg]

Don't forget the filter cap.

I didn't. 10uF is more than enough.

I'm with you about the divider values. Sometimes if you don't know how to do it look how the "big ones" did it and just copy it although they also make mistakes.

It use 470k/82k and not 470k/820 ohm.

It's 470k/82k (R81/R83) in the JVM410 schematic.

[catalin gramada]

https://ohmslawcalculator.com/voltag...der-calculator



Schematic (JVM410) was posted above. It doesn't get more simple than that.
Sometimes simple things here are getting unnecessarily complicated and drowned theory.
In this case take the HV from screens node. Calculate elevation volts (top resistor 470k-1М)). Connect heaters' CT or artificial CT to voltage divider's node. Done.

Don't forget the filter cap...
Some basic teory still necessary to understand what you're doing.
First, choose a supply node from at least two point of view : inevitably the divider will draw some current to ground (the divider is a power consumer by itself), and we don't want to affect the voltages over the chain supply so is preferable to install as close to the power supply and not in the end of the supply chain. Second: we want a quiet supply node for our DC reference, as much free of ripple. From this point the screen supply node is good compromise. Calculate the voltage you need . Choose the values of resistors in respect with voltage you need but also how much current you want to spend. But keep in mind a large value of series resistor, like 470k-1M will drastically limit the charging current of the filter capacitor, meant you will get the nominal voltage in seconds or tens of seconds. From the same reason the filter cap should not be larger than 10-20 uF large enough to smooth the ripple for our new filament reference point. Is all compromise.(for instance: 1M resistor with 100uF cap have a time constant 100 sec.)-you have a parallel resistor here meant bigger series resistor=longer time to stabilise.
From the screen node put a 220k series with 22k shunt. Instal a 10-22uF/ 100v cap over shunt resistor and tie the filament center tap to this . Done.
Eg: Suppose you have 380v at screens 220k series/22k shunt=34.5 V , 220+22= 244k, 380/244000=1.56mA , 380-34.5=345.5V over series resistor. 345Vx0.00156A= 0.538W power dissipation. 1W resistor will be enough, but 2w better choice.
From a 440V supply node with 220k/22k you get 40V DC elevation. Just change the values in the picture you posted: replace 470k with 220k/2W and 820 ohm with 22K.
470k/820 do almost nothing. Is a minimal elevation maybe to suppress some noise due h-k coupling but is not a elevation in a sense we used to protect heater to cathodes insulation at high potential like in cathode folowers or cathodyne inverter circuits. You forgot to tell us what is the reason you intend to elevate the heaters reference? We usually elevate the heater reference for a reason. And as the side aspect- you may find slight difference in sound.
But could be a mistake as well, as we see in jvm 410 schematic posted above. It use 470k/82k and not 470k/820 ohm. They made it in a rush and was confused? You have to take time for everything when ask two grands for an amp...I think. 820 ohm is not the same as 82K of course . But maybe some of those mistakes who made a difference from a boring to a brilliant amp? It is as simple as hell...

[Gregg]

Could you tell me how to calculate the resistor values to use for the divider? Ohm's law?

https://ohmslawcalculator.com/voltag...der-calculator

Let's have a schematic please.

Schematic (JVM410) was posted above. It doesn't get more simple than that.
Sometimes simple things here are getting unnecessarily complicated and drowned theory.
In this case take the HV from screens node. Calculate elevation volts (top resistor 470k-1М)). Connect heaters' CT or artificial CT to voltage divider's node. Done.

[Chuck H]

I'm with Helmholtz. Let's have a schematic please. I didn't peruse the Valve Wizard link (though I'm sure it's informative) but it's not hard to do. So since you have questions I think it would be best if you post a schematic and take instruction. Since this seems to me a new build I'd be inclined to just integrate the divider for the filament elevation in the HV bleed circuit.