Bell & Howell Filmosound 179

[xtian]

A little chopsticking made it obvious that I'd have to reroute the OT primaries away from the PI. Done. And this made a BIG reduction in the amount of hum.

Now I notice that if I ground the input (pin 1) to the PI, the overall noise is quite acceptable. So I'm getting induced hum or ground loop hum in the first stage. I'll try rearranging my grounding scheme next, based on Mark's comments.

[xtian]

Well, I moved a lot of stuff around, making improvements in lead dress and grounding scheme, but improving the noise only very slightly. It's not too bad right now, though. Wouldn't seem out of place on stage, where I've heard some very noisy amps at rest. (Right?!? Sheesh.)

There are two more steps I can take to reduce hum: isolate the output jacks from the chassis, and elevate the heater CT.

what about elevating the heaters off B+ to about 70VDC?

How would you do that?

Anyway, I plugged the Filmosound into my 2x10 cab and gave it a work out with my Tele. Very, very cool. LOTS of volume and gain on tap. In fact, I may have to quiet it up a little. Was getting some squeal or oscillation with controls dimed.

Where's the most likely place to reduce gain? I'm thinking about the four resistors that supply DC to the plates of the preamp and PI. On the schematic they're spec'd as 250K but I used 220K.

Thanks for the help, peeps! I'll post video when I get everything tweaked.

[surfsup]

surfsup wrote: what about elevating the heaters off B+ to about 70VDC?

xtian: How would you do that?

A voltage divider off a B+ node, with a cap to ground in parallel with the resistor to ground. So two resistors and a cap. Let's say you have 350V at B+, if you used a 180k/47k combo that would be:

350*(47k)/(47k+180k)=72V

Put the 180k off your B+ to a strip or eyelet, the 47k from the 180k to ground. Put a 100u/125V cap across the 47k to ground, and run a lead from your heater CT or 100R/100R to the eyelet between the 180/47.

Kinda cramped in there though.

[Phil_S]

A little chopsticking made it obvious that I'd have to reroute the OT primaries away from the PI. Done. And this made a BIG reduction in the amount of hum.

Is the picture filmosound05.jpg showing that? I can see you've relocated the B+ wire from fc#1 (fc= filter cap) into the fold by the open side of the chassis, but that's all I see for the 6v6 that is under the jack farthest from the fuse. What I see there is the OT primary to pin 3 is jumping over the filament wires and the blue wire from the PI cathode running over it. Even though you have observed the right angle rule, that alone may not be enough and may be a hum source. (It is difficult to know; I am speculating.)

Whether the suggestion that follows is worthwhile, I am not sure. If it was my amp, I'd be willing to give it a try.

You've got some room on the brown board. Move the cathode R/C from the PI down two positions (towards the input jack and pots). Actually, you only need to move the + side; you can leave the ground side where it is, but I think it is clipped too short to reach, so you might improvise a short extension lead. Attach the OT primary to the 3rd lug from the top, where the R/C is now. I am thinking, if you are lucky, it will reach. Then run a jumper to pin 3 of the tube socket. This gets it away from the filament wires. It also gets it away from the blue cathode wire from the PI to that R/C.

On another matter, are your output jacks grounded or floating? I see a bridge between them, but I don't find a ground wire. It would be easy to miss -- pictures can be hard to read. Assuming it's not there, this amp might benefit from a ground to the - side of fc #1, as Mark suggests. You've got a great spot to run a wire from the jack closer to the fuse, under the right edge of the board next to the power tube cathode R/C and snake it around to the filter cap ground. I am not seeing the OT secondary connection for the ground/sleeve, so this whole comment on the sleeve ground may be worthless.

More on the grounds...I'm not fond of the spot where the power tube cathode R/C is grounded. It is tied to the ground for fc #3 which is supplying the PI and the preamp. This is mixing ground stages. I think it is better the keep the ground stages separate. I normally ground the power tube cathode to the same spot as fc #1. There are spare pins on the unused socket to help you with this.

I think, if you keep at it, results will happen.

[xtian]

Excellent advice, guys! Thank you. I'll keep hacking on it.

Any comments on how best to reduce gain in the pre and PI stages?

Where's the most likely place to reduce gain? I'm thinking about the four resistors that supply DC to the plates of the preamp and PI. On the schematic they're spec'd as 250K but I used 220K.

[Phil_S]

To lower the gain on the input tube, lower the value of the plate load resistor and/or increase the value of the cathode resistor, and/or lower the value of the grid leak resistor.

If built to the schematic, you should have for V1, 2.5M grid leak, no Rk, and Ra 250K. So, you can't lower Rk

For the PI, pretty much the same routine. I don't know if I'd go down much from the 220K plate load value. If you are using 2.5K on the cathode, you might consider reducing Rk. The grid leak is really the volume pot, so maybe you don't want to mess with that, or use a 500K instead of 1M? I'm not sure what a change in pot will bring you because you can already make it equal to 0!

See chart #5 in the attached. You can improvise from the chart.

BTW, I should tell you how much I like this build. It is no small task to shoehorn an amp into a small chassis and keep the price low, too. This is "my kind of build." I think it is relatively easy to get a good result when you have unlimited real estate inside the chassis. Where's the fun factor in that? (OK, that's fun, too, but not the same kind of fun. I think you know what I mean.)

[xtian]

Well, hell. I have less and less trust in "proper" grounding schemes. I went to the effort of isolating the output jacks and routing a ground back to the PT star ground, and it made NO difference at all.

I went even further: I lifted the input jack star ground point, connected an alligator lead, and went around merrily grounding it to every point on the chassis, figuring I'd hear, at the very least, some difference in hum or noise. NOTHING! All ground points seem equally good or bad.

Move the cathode R/C from the PI down two positions (towards the input jack and pots). Actually, you only need to move the + side; you can leave the ground side where it is, but I think it is clipped too short to reach, so you might improvise a short extension lead. Attach the OT primary to the 3rd lug from the top, where the R/C is now. I am thinking, if you are lucky, it will reach. Then run a jumper to pin 3 of the tube socket. This gets it away from the filament wires. It also gets it away from the blue cathode wire from the PI to that R/C.

All done. Looks prettier, sounds the same. I mean, it sounds fine, just not great. If I was At Mars, I couldn't sell it for $900 and keep a straight face…OK, even if my amp was dead quiet, I couldn't sell it for $900 and keep a straight face.

[Cliff Schecht]

I wouldn't pay $900 for that. I have amps that people would say the same thing about if they saw the insides. But when they hear them, they try to buy 'em anyways!

[Phil_S]

It's going to be a head scratcher. Patience is required here, because there is probably one thing that will fix it. Since we don't know what, it becomes trial and error.

One of the new pictures shows be the B+ supply for the OT -- the wire with the green heatshrink. I couldn't see this before the way this pic shows it. It comes really close to the secondary connection that was just moved onto the board to get away from the filament wires. It also comes close to the filament wires. I'd urge you to re-route that one. If this were mine, I wouldn't want to disturb the end that's connected to the OT as it looks as if it is already very short. I might think about lifting the wire at the filter cap. Because of what I see under the brown board -- packed tight, I might try to pass that wire out the other side of the brown board, down towards the input jack (but not too close!) and then along the front side of the amp, up onto the red board and around the fuse side of the cap. I don't think I'd want it near the pots. I'd go for that "tunnel" between the PI and unused socket and under the R/C for the power tube cathodes.

Have you separated that power tube R/C ground and moved it to the first filter cap ground?

The latter is simple, so I'd do that first. IMO, both of these carry potential, but, as you have already learned, seemingly good thoughts can produce nothing. It's really important to do only one thing at a time. If you do multiple things that makes it worse, you won't know which one to put back.

Oooh boy, I thought to take one last look at the new pics. Do I see the mains (green/yellow) ground sharing it's ground point with the circuit ground? IMO, that is a sure fire way to introduce 60 Hz or other hash from the house wiring. It is also unsafe because now your circuit is directly connected to the mains, bypassing the fuse and the PT. You must give the mains ground its own private ground point. I'm not a fan of using the PT bolt as a ground and certainly not for the mains ground. Although maybe not the best, it looks to me like the strain relief screw might be OK for the job on a temporary basis. Please fix this, even if you do nothing else.

The proper way to deploy the mains ground is with a ring lug, so it can't get pulled off its bolt. Double lock it with a nut and star washer to secure the bolt, star washer (unless you use a toothed ring lug), ring lug, and another nut. Make it good 'n' tight, but don't break the bolt! There should be adequate slack in the ground wire so, in the event of disaster while plugged in, it is the last thing to break free from the chassis.

[surfsup]

Your 6v6 cathode ground is connected to your pi ground on the "circular" terminal strip. What do you guys think about separating the yellow leads from the tube socket terminal strip and run them to the ground turret for the second cap?

Your neg feedback, grn led, looks to run around the whole amp full circle! Going to the 1 meg on he board then circling back using he blue lead? Couldn't hat be shortened by running the resistor from the Jack to the power tube socket directly?

[Phil_S]

Your 6v6 cathode ground is connected to your pi ground on the "circular" terminal strip. What do you guys think about separating the yellow leads from the tube socket terminal strip and run them to the ground turret for the second cap?

That makes two of us. The 1st and 2nd cap are grounded together. I said 1st, you said 2nd, but it is the same.

Your neg feedback, grn led, looks to run around the whole amp full circle! Going to the 1 meg on he board then circling back using he blue lead? Couldn't hat be shortened by running the resistor from the Jack to the power tube socket directly?

Good call, but it only works if the phase is correct. He's got to pick the right output tube. Yes, go directly from the tip of the output jack to the grid of the 6v6.

I feel certain this amp can be a real winner. I'm rooting for you Xtian.

[surfsup]

Hmm I missed your comment about that...I agree its tiiiieeeeght in there! Hope he works it out.

[M Fowler]

With proper voltages and grounding there is no reason this amp can't be dead quiet.

My 5b6 basically same amp and tubes is so quiet the only way you can tell the amp is on is the pilot light or hit the strings on the guitar.

[xtian]

My 5b6 basically same amp and tubes is so quiet the only way you can tell the amp is on is the pilot light or hit the strings on the guitar.

Yeah? Well, now you're just bragging.

I appreciate all the great advice, guys. I'll get back into it soon.

Anyone have a metal fab shop? I need a pair of handles built.

[Phil_S]

With proper voltages and grounding there is no reason this amp can't be dead quiet.

My 5b6 basically same amp and tubes is so quiet the only way you can tell the amp is on is the pilot light or hit the strings on the guitar.

Mark, The space in your chassis looks like the great plains compared to what Xtian is working with. He needs a shoehorn to get into his. Even so, I am in agreement, this amp can be made quiet.