Channel Switching JCM800 Design? + BIAS pot question

[ViperDoc]

I have a question about the placement of the bias adjustment pot on this 800 amp. See the following photo:



I have a 50KL shaft down CTS pot on the left, far away from the board (see standoff’s). I also have an internally adjustable 10KL bias pot closer to the board; these are just here for mocking using the pre-drilled chassis holes. I hate the idea of having to pull the chassis to rebias an amp. I’ve used a Loknob on previous builds to protect the adjustment shaft and they work great. The PT is right underneath the closer location, so I can't do the 50K pot there. Question is, how long can the wires run in the bias adjustment before it becomes a problem? Obviously the outer location would be much easier to access and adjust, but how far away can you really go? Thanks.

[pdf64]

Yikes, a 50k adjustable resistor would allow the bias voltage magnitude to be turned down way too much, such a degree of variability is bad.
But that arrangement is a retrograde change from the original Marshall arrangement, see R26 and RV1 https://el34world.com/charts/Schematics ... w_2204.pdf
Because if the pot wiper fails, the power tubes will bias cold, so fail safe.
Whereas with the Mojo arrangement, if the pot fails (eg wiper is on a bad bit of track) the power tubes will lose bias and enter melt down. Yes I acknowledge that's the way BF etc Fenders are, but why change a design so as to facilitate such a potentially damaging failure mode?

Whatever, the length of the wires shouldn't matter much as it's a low impedance circuit, fully bypassed to 0V with the Marshall, whereas the bias supply output with the Mojo arrangement may be a few kohms, which probably isn't significant.

[martin manning]

The supplied 10k bias pot is accessible from outside the chassis, through it's bushing. Is the transformer bell obstructing it?

[ViperDoc]

The supplied 10k bias pot is accessible from outside the chassis, through it's bushing. Is the transformer bell obstructing it?

Yes, the PT is right in the way!!! You can see the corner of the PT housing through the other hole just above where the 10KL pot is. So I can access the pot through the bushing, that's good. How do I do that while the amp is on without electrocuting myself? Just seems like a PITA to get to either way.

[ViperDoc]

Yikes, a 50k adjustable resistor would allow the bias voltage magnitude to be turned down way too much, such a degree of variability is bad.
But that arrangement is a retrograde change from the original Marshall arrangement, see R26 and RV1 https://el34world.com/charts/Schematics ... w_2204.pdf
Because if the pot wiper fails, the power tubes will bias cold, so fail safe.
Whereas with the Mojo arrangement, if the pot fails (eg wiper is on a bad bit of track) the power tubes will lose bias and enter melt down. Yes I acknowledge that's the way BF etc Fenders are, but why change a design so as to facilitate such a potentially damaging failure mode?

Whatever, the length of the wires shouldn't matter much as it's a low impedance circuit, fully bypassed to 0V with the Marshall, whereas the bias supply output with the Mojo arrangement may be a few kohms, which probably isn't significant.

I see. Thanks for the reference! A few points:

1) This is the first diagram I've seen that shows the bias resistor to ground on the pot itself. You're saying if the pot fails, there's no backup and my tubes Chernobyl in mighty fashion. Would you recommend a different arrangement?

2) I considered a 50KL bias pot because I have a few of them, and that's the value that was spec'd on my JTM45 build to allow for KT66. This JCM800/Plexi schematic denotes options for wired-in bias resistor values for EL34/KT66 that differ by 30-ish K, so if I bias the circuit correctly with the series pot configuration, will this be OK?

Thank you!

[martin manning]

There is not much voltage there, but I usually use a bamboo skewer that has one end shaped to look like a screwdriver. That might be skinny enough to get by the PT bell. I agree it’s nice to bias with the chassis in the head box, but you need some way to measure current too. I would add 1 ohm 1% resistors to ground the power tube cathodes to make that easy, then you need access to those points (pin 8’s) from outside.

[ViperDoc]

There is not much voltage there, but I usually use a bamboo skewer that has one end shaped to look like a screwdriver. That might be skinny enough to get by the PT bell. I agree it’s nice to bias with the chassis in the head box, but you need some way to measure current too. I would add 1 ohm 1% resistors to ground the power tube cathodes to make that easy, then you need access to those points (pin 8’s) from outside.

Thanks, Martin. I have a TAD Bias master, is that sufficient?

[pdf64]

1 - the arrangement in the previously linked Marshall 2204 schematic is 'fail safe' and should offer a good range of adjustment.
2 - Really need to see a sketch of exactly what circuit you're intending for the bias supply circuit and adjustment; just hand draw it out, take a photo and upload it somewhere, provide a link.

[ViperDoc]

1 - the arrangement in the previously linked Marshall 2204 schematic is 'fail safe' and should offer a good range of adjustment.
2 - Really need to see a sketch of exactly what circuit you're intending for the bias supply circuit and adjustment; just hand draw it out, take a photo and upload it somewhere, provide a link.

Thanks!

1) The 2204 schematic shows a resistor leading into the center lug of the adjustment pot, which is different than the mojo wiring diagram/schematic. Your schematic appears safer, and is more in line with what I've seen in the past. Seems better.
2) Please, see the schematic in the first post. Lower right, you'll see a 47K for EL34 and an 82K for KT66 (approx., I'm sure), but based on that, the 50KL pot allows for more than enough range without changing the wired in resistor, yes? I'm lazy, man. If I can wake up one day and feel the need for KT66 knowing it's a tube change and a pot twirl away, I'm good! But is this bad? I'm open to suggestions, that's for sure.

I'm concerned about the pot leading into the resistor. Should I be? I'm still learning. Thanks, man!

[pdf64]

To accommodate EL34 and KT66, I suggest to get rid of the 10k pot, and wire up a 47k resistor in series with the 50k pot, wired as a variable resistor, as per R26 and RV1 of the 2204 https://el34world.com/charts/Schematics ... w_2204.pdf

[ViperDoc]

To accommodate EL34 and KT66, I suggest to get rid of the 10k pot, and wire up a 47k resistor in series with the 50k pot, wired as a variable resistor, as per R26 and RV1 of the 2204 https://el34world.com/charts/Schematics ... w_2204.pdf

OK. I find the mojo wiring confusing. I see that the 2204 schematic has the bias resistor R26 and pot RV1 in series to ground, and those are, together, in parallel with C19. In the mojo diagram, the bias wire comes into the center of the bias pot, the bias resistor is connected to the ground lug to ground (series connection), and the other wire is tied to the 8 uF cap in the bias area. This to me looks like a series connection in regards to the bias cap. (I don't even know if that's what it's called, sorry man.) The JTM45 I just built had the bias resistor and bias pot in series and in true parallel to the bias cap using the same turrets. It's hard for me to see whether the diagram conforms with the 2204 schematic electrically. I've made some changes to my board to accommodate the new diode arrangement, so I'll sketch up how I might reconfigure it and pass it by you. Meanwhile, mojo updated the bias on their JTM45 diagram as follows:



This looks more like what you recommended.

[ViperDoc]

I sketched up a reconfiguration of the JCM bias supply and the diode arrangement some of you kindly recommended. I appreciate any comment you might have on this. Thanks in advance. I added the turrets in "purple" to try to improve the arrangement and prevent airborne connections between electrolytics.

[pdf64]

That looks good. Nice drawing, what software are you using?
You may prefer to swap around the yellow and orange wires on the pot terminals, depending on whether you want the idle plate / cathode current to increase or decrease as the pot is turned clockwise; I think that the way you've got it, the current will decrease (ie as the bias voltage increases, the plate current will decrease).

It's great the Mojo have amended the standby on their JTM45 kit layout to stop it being 'hot switching'.

Regarding the comment I made in another thread of yours

Note that the wire from the CT of the HT winding is the noisiest of an amp, as it carries large current spikes required to charge the reservoir cap; hence it should connect directly to the -ve terminal of the reservoir cap, and be kept well away from sensitive circuits. Avoid making seperate connections of these to a common 0V point / star ground, as the current spikes may contaminate other circuits also connected to that common point.

using that latest Mojo JTM45 layout, the HT CT red/yellow wire from the fuse to the ground tab should rather go to the -ve terminal of the reservoir cap, to the left of the PT. That way the currents circulating around that ground tab will be 'cleaner', as the most significant current spikes have been routed away from it.

[ViperDoc]

That looks good. Nice drawing, what software are you using?
You may prefer to swap around the yellow and orange wires on the pot terminals, depending on whether you want the idle plate / cathode current to increase or decrease as the pot is turned clockwise; I think that the way you've got it, the current will decrease (ie as the bias voltage increases, the plate current will decrease).

It's great the Mojo have amended the standby on their JTM45 kit layout to stop it being 'hot switching'.

Regarding the comment I made in another thread of yours

Note that the wire from the CT of the HT winding is the noisiest of an amp, as it carries large current spikes required to charge the reservoir cap; hence it should connect directly to the -ve terminal of the reservoir cap, and be kept well away from sensitive circuits. Avoid making seperate connections of these to a common 0V point / star ground, as the current spikes may contaminate other circuits also connected to that common point.

using that latest Mojo JTM45 layout, the HT CT red/yellow wire from the fuse to the ground tab should rather go to the -ve terminal of the reservoir cap, to the left of the PT. That way the currents circulating around that ground tab will be 'cleaner', as the most significant current spikes have been routed away from it.

Much appreciated, pdf! I’ve been using DIYLC to draw up circuits. It is free to download, and while the download file usually is interpreted as “corrupted”, it only requires a legacy java runtime in order to function properly. It works like a charm.

You are correct, I believe, regarding the bias pot polarity. I learned this the hard way when I found my KT66 tubes at 120 mA when I first measured it! I’ll take your advice and reverse it.

On a separate note, how would you properly wire individual bias pots in this circuit?

[pdf64]

Thanks for the DIYLC tip, I've downloaded it before but shied away at first use, when it wasn't immediately intuitive
I'll grit my teeth, knuckle down and try again

Regarding more bias outputs, I'd use Merlin's 'full wave' arrangement (may need to tinker with the R1 value), and have C1 supply parallel sets of the P1, R2, R3 and C2 arrangement http://www.valvewizard.co.uk/bias.html