• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Understanding 845 PP Bias

Can someone help me understand the Bias of the Marantz T1?
This first schematic is the "Musicians Amplifier" from Lilenthal Engineering website.
http://bit.ly/2OnrQnv

[​IMG]

Now I understand how the Bias is set here, it's easy to see and understand.

These next three schematic are from Sound Practices issue #1
The 1936 UTC amplifier is also pretty easy to see and understand the Bias set up. With the two pots coming off the LS-6 secondary.

[​IMG]

Next is the MARANTZ T1

[​IMG]
This amplifier is where I get lost in understanding its Bias control.
It's my understanding that the Heaters on all three stages are fed from DC power supplies. Not shown here.
I've looked for a while now and this one schematic is pretty much all there is on the internet.
Be that as it may. How is the Bias adjusted here? Or is it fixed? Or might it be something that is adjustable, its just that the part of the circuit where Bias control is set is not shown here, so we have to imagine a control circuit elsewhere?

And lastly is a circuit from MJ magazine with the 845's being driven from a single Interstage transformer.
Again same question.
How would Bias be adjusted here? It looks like there are two separate filament transformers here.
And I'm assuming that the Bias section just isn't shown.
The power supply section is obviously missing so again the Bias would probably be a part of that circuit.

I'm wondering if a Bias scheme similar to the one used in the "Musicians Amplifier" would be applicable here ???
 

Attachments

  • Musicians-amplifier-Senior-845-PP-Sarser-Sprinkle-1951-ed.jpg
    Musicians-amplifier-Senior-845-PP-Sarser-Sprinkle-1951-ed.jpg
    407.8 KB · Views: 576
  • dc208427835fb1801da1c8f340b82456.jpg
    dc208427835fb1801da1c8f340b82456.jpg
    131.8 KB · Views: 404
  • T1.gif
    T1.gif
    30.3 KB · Views: 407
  • 923b34df45d8a7c2fb9753d958634f82.jpg
    923b34df45d8a7c2fb9753d958634f82.jpg
    93.2 KB · Views: 397
  • 314b9909b3397d5e0fc38ada72752a79.jpg
    314b9909b3397d5e0fc38ada72752a79.jpg
    154.3 KB · Views: 401
Last edited:
The Marantz T1 is an interesting circuit. But it's NOT the amp I would have built.
First of all I would exchange the UTC LS-6L4 for a FREED QGA-42 output transformer. The Freed is rated for 100 watts @ 5K ohms.
Secondly I would not use 845 for the HV rectifiers. 836, or 866A or even solid state.
Third bump the B+ up to 1100 or 1200 volts so 80 to 100 watts out is feasable.
Of course that means beefing up the cap ratings too.
 

6A3sUMMER

Member
2016-06-07 6:50 am
Left most schematic
915V plate, 85V drop across the self bias resistors
A 500 Ohm bias rheostat resistor and 200 Ohm balance resistor allows setting of the total current, and balance, respectively of the two 845s. There are two more resistors (*) to get the total resistance of the self bias circuit so the range of current is correct for the actual B+ voltage.

Next schematic (UTC)
Adjustable Individual Fixed Grid Bias.
Unfortunately nobody lived to tell of the bias adjustment. The two 1/4 inch jacks are for plate current meters, but each one has the 900V to 1000V on each jack sleeve and nut (how shocking). You need to enclose those connections away from hands, and the meter leads too.
Also, you need to be sure the 83 mercury vapor rectifier for the negative grid bias is powered and up and running before the B+ comes up, or you get meltdown of the 845s. Actually, the 83 filament secondary and negative voltage secondary are on the same transformer (not good, supposed to warm up mercury vapor filaments before the plates are rectifying).

Marantz T1
The two 1.5k Ohm self bias resistors and 22 Ohm current sense resistors (TP 5 and 6), are the self bias network. Each series network of 1.5k and 22 Ohm are bypassed by a capacitor.
Individual self bias, nice.

The next schematic is simply the old print, and the first schematic is the same Senior Amp.

The last schematic is incomplete, and so no conclusion is possible.

I favor using individual self bias networks, (resistor and bypass cap). And that requires individual filament supplies.

I recommend DC filaments.
With AC filaments you get 2x line frequency sidebands on each musical note / tone.
That is true, even if you balance out the hum (like no hum when there is no music playing).

Do you really need 80 to 100 Watts?

This is a dangerous and very demanding build.

Safety first.

"You should make it as simple as possible, but no simpler" - Albert Einstein
 
Use the powersupply from Sakuma's amp with the 6J5 driver circuit.
The Bias circuit is what I was looking for.

Probably sub out the rectifiers with 836 or 866A's
And maybe use 6BL7GTA instead of 6BX7 ??? A little R&D needed there.
I had Phil at Heyboer make me the interstage transformer labeled NC-9 it's 5K to 20K ohms ct.
And use FREED QGA-42 output transformer. 5K ohms at 100 watts.
 

Attachments

  • 923b34df45d8a7c2fb9753d958634f82.jpg
    923b34df45d8a7c2fb9753d958634f82.jpg
    93.2 KB · Views: 118
  • 845pp2.gif
    845pp2.gif
    6.5 KB · Views: 111
Last edited:

6A3sUMMER

Member
2016-06-07 6:50 am
Your post # 6 has an 8 uf capacitor to provide all the plate current for two 845 tubes.
8uF is 995 Ohms at 20 Hz.
You will only be able to operate in class A on the output stage. Otherwise, the 8uF cap will cause the B+ to drag down, and clip the output stage.
You should use more capacitance than 8uf.

As to bias circuits, first decide whether you want fixed adjustable bias, or self bias.
For interstage driver secondaries that do not have two separate windings, and using fixed bias, you can not individually adjust the current in the two 845 tubes.

Individual filament supplies, individual self bias resistors, and individual bypass caps . . .
many tubes will have similar plate currents. And self bias helps prevent thermal run-away.

To determine the bias voltage, use the 845 tube curves, and determine the plate to filament volts you want to use, and the current you want to run at.
Your bias circuit will need to provide that voltage (fixed or self bais).
 
Last edited:
Just my personal opinion of course, and not a criticism, but 5K Ohms anode-to-anode is a poor choice for *any* of the 50-watter valves, even the least stringent type 845. Twice that puts the valves into the linear category and twice that again gets into the truly magic realm of great linearity and decent loudspeaker damping. Driving an actual loudspeaker is, after all, the goal.


These historical designs are interesting as historical items, but aren't even remotely carved into stone tablets and carried down from mountaintops. Push-pull type 845s is an exciting *new* thing, to be made the best of. Rant mode off, sorry.


All good fortune,
Chris
 

6A3sUMMER

Member
2016-06-07 6:50 am
In post # 10, I thought the bypass capacitor of 12 uF was too small for the 2.4k self bias resistor and the 845 tube. So I decided to find the amount of capacitance that I would use there.

Suppose the plate load was 5k Ohms and 40 Henry inductance. 40 H is 5000 Ohms at 20 Hz.
The 845 plate sees 2500 Ohms at 20 Hz.

The 845 u = 5.3, and its transconductance = 3100 micromhos.
The “cathode” impedance at 20 Hz is 2500 Ohms/(u + 1) in series with 1/3100umhos.
2500/6.3 = 397 Ohms. 1/3100umhos = 323 Ohms. 397 + 323 = 720 Ohms

2400 Ohms in parallel with 720 Ohms is 554 Ohms.

When the difference of the ac voltage at the “cathode” is different when the bias resistor is unbypassed (unity) versus bypassed (0.1 x unity), the frequency response will be -1 dB there.

If we want the frequency response due to the bypass cap to be -1 dB at 20 Hz, we need the capacitive reactance to be 0.1 x 554 Ohms = 55 Ohms.

Capacitive reactance, Xc = 1/(2 * pi * f * c)

1/c = 55 * 2 * pi * 20 Hz.

c = 145 uf. We can use a 150 or 200 uF cap there and be - 1dB @ 20 Hz.

In my opinion, a 12uF bypass cap there would cause severe low frequency roll off.

Why your mileage may vary:
If the plate load was different than 5000 Ohms and the inductance was different than 40H, the bypass capacitor would need to be adjusted to a different value in order to be -1 dB at 20 Hz.