HoSo56 Single Ended KT66, 2x5879, Ghetto UL

[DummyLoad]

Thank you all for your responses. I chopsticked a couple of signal wires vs the heaters. No difference. I resoldered the preamp caps ground wire. No change. Still humming, still getting scratching along the bussbar.

So I pulled the relay out of the relay board. Presto change-O! Quiet as a church mouse peeing on a cotton boll.

is the relay power supply grounded? did you try it with ground floating? is the foot-pedal jack for the relay (assuming you have one) an insulted jack? relay power ground and signal power ground should not be tied to the same point.

--pete

[foghornleghorn]

Relay PS is grounded only thru the footswitch jack on the chassis. The footswitch jack is not insulated from the chassis. There is no signal ground at that point, or anywhere near it. The relay PS is powered by the 5V rectifier winding on the PT, which is not center tapped.

Unless the relay is defective, which I have not yet checked, I suspect maybe the relay wiring. The NO relay contacts are connected to the 1uF cap and 390pF cap, the NC contacts are not connected to anything and the Common contacts are connected to the treble pot and the top of the 1K5 cathode resistor on V2A. This is how I read the schematic.

When the footswitch is plugged in it does not connect thru to ground in either position so I probably have miswired it.

[foghornleghorn]

There is a discrepancy between the layout and the schematic re the wiring of the 390pF cap. The layout is I think wrong. I rewired it to the schematic but no help on the hum/buzz.

I insulated the FSW jack from the chassis and took the ground (barrel) wire to the relay ground which goes to the sole preamp chassis ground.

I also attached the relay PS 7805 regulator to the chassis with a nylon screw and nut.

I also fixed the FSW wiring. I had the FSW LED in backwards.

Currently with the relay
removed from the relay board the amp works fine, no hum, just some hiss when it's cranked, which I would expect. With the relay back in, hum/buzz is back, with the boost engaged, the hum/buzz is worse.

Update. Thanx to help from Dave Cohrs of Colossal Amps, I think once I get the relay board rewired to remove all DC from the relay contacts hopefully that will fix the problem. This entails rerouting the 1uF boost cap so that the negative is connected to the relay common and it grounds in the NO connected mode ie relay engaged. See revised schem and layout dated 9-29-13.

[foghornleghorn]

Rewired the 1uF cap to get DC off the relay Common. No change, still hums/buzzes with relay in, works perfectly OK with the relay pulled out.

If I had any hair left I'd be pulling it out right now.   

[sluckey]

Where is the schematic for the relay power supply??? That's what we need to see.

[DummyLoad]

pretty sure the problem is with the 390pF cap switching in the TS. try disconnecting all of that circuit. simple cathode bypass cap >> gnd switching shouldn't do anything but pop when switched with arrangement shown.

--pete

[foghornleghorn]

Thanx for the responses Gents. The 390pF/1uF switched caps is right off the original HoSo56 schematic, I just modified it to keep the DC off the relay.

Not sure I have a schem for the relay PS board, I bought it preassembled from Dave Cohrs of Colossal Amps. I've used several of them before without problems.  It is putting out 5.0V DC OK, fed off the 5VAC PT winding. The relay board itself is built by him too, have also used them before without any problems.

I'll try deleting the 390pF cap and see what happens.

[foghornleghorn]

This just keeps getting weirder. I disconnected both caps from the relay. Still hums. I disconnected the footswitch jack. Still hums. I then disconnected the 5VAC winding from the relay PS. Still hums.

Before, it would hum with the relay in the board but NOT with the relay out. Now it hums with the entire relay circuit disconnected.

So it can't be anything in the relay/FSW circuit because none of it is in circuit.

The preamp caps are grounded to the buss bar. The power amp caps, KT66 cathode resistor/cap and the HT center tap are all grounded to the chassis near the AC inlet.

I think I have to check my cap ground joints.

[Willabe]

The preamp caps are grounded to the buss bar. The power amp caps, KT66 cathode resistor/cap and the HT center tap are all grounded to the chassis near the AC inlet.

I know there's disagreement on this but the PT B+ CT, heater CT, power amp caps and KR have the highest current flowing to ground. The input jack has the lowest signal in the amp other than a reverb tank return.

Power grounds should be kept away from preamp/low signal grounds.


            Brad     

[foghornleghorn]

Looks like I may have misled you. By "AC inlet" I meant the wall AC, not the "input jack". Only the preamp caps are grounded near the input jack, via the bussbar on the pots. Everything else in the main PS is grounded at the power supply end of the amp.

[HotBluePlates]

Can you pull a preamp tube at a time to see where the hum is coming from?

Unrelated question: Why do you have a big 5.1kΩ screen resistor on the output tube? That would have to be hurting UL output...

[foghornleghorn]

Pulling tubes was the first thing I did, but I've not done it since I made all these changes. I'll try it again. IIRC pulling V1 made no change, pulling V2 killed the hum.

The 5K1 screen gridstopper was intended to lower the screen voltage a bit and maybe add some compression. I know that in audio this would be 220r even if it was there at all. Would you recommend I reduce it and if so to what value and why?

I am now going to try three things:

1) remove the relay board from chassis ground-everything else is disconnected.

2) take the preamp caps ground from the buss bar to a new chassis ground to the left of the buss bar.

3) reconnect the relay boardand FSW ground to chassis ground at the power amp caps/HT CT/power tube cathode bias ground point.

[foghornleghorn]

Confirming removal of V1, no change. On V2, just the capacitive effect of grasping the tube to pull it out greatly reduced the hum. With V2 out, hum disappeared.

1) Did not change anything, BUT 2) relocating the preamp caps ground off the bussbar, to a new, separate chassis ground WORKED 100%!! No hum/buzz at all. Plays great again. Finally. Lesson learned!

Now to gradually reconnect the relay/FSW circuit and see what happens. I will relocate the relay ground away from the preamp chassis ground too.

My thanx to all who offered help along the way!

[foghornleghorn]

Reconnected the relay circuit, works fine, no pop on switching, BUT the 390pF cap had to go, it caused the hum to come back.

When cranked this amp has lots of sustain, sounds almost Dumble-y sometimes.

I'm going to put in a speaker impedance switch too, and rewire the OT secondary so I can use it at 4 ohms with both my 2x12 Celestion blackback cabs. This will change the OT primary a little at 4 ohms, to 2K8. Will still be 3K at 8 ohms.

The 1uF cap in on V2A doesn't boost a whole lot, about 6dB according to TubeCad; not like switching out the mid pot ground would.

Now to tweak a little. I think I upped the TS slope too much. Will drop that from 100K to 51K, that is still higher than the original HoSo56 which is 33K. Will probably play with the screen gridstopper on the KT66, and may try a gridstopper on V3, the driver. V3 plate might also benefit from a smoothing cap across the plate resistor.

Any other suggestions? Have already drawn up a new schem with two KT88s in parallel, being driven by a 6U8A/6GH8A pentode with the triode as a CF to the KT88s, as these are tougher to drive than KT66. Am I pushing my luck here?Naturally I'd need bigger iron.

[HotBluePlates]

The 5K1 screen gridstopper was intended to lower the screen voltage a bit and maybe add some compression. I know that in audio this would be 220r even if it was there at all. Would you recommend I reduce it and if so to what value and why? ...

You can have a squishy, compressed amp or you can have an ultralinear amp; in general, you can't have both.

Also, in UL you assume the screen voltage is same (or slightly higher) as plate voltage, by virtue of the connection to the OT.

Already, SE UL is a rare bird. I recommend getting it working without trickery, if for no other reason than you'll understand what makes it sound the way it does.

... relocating the preamp caps ground off the bussbar, to a new, separate chassis ground WORKED 100%!! No hum/buzz at all. ...

Was each ground for the preamp (fed by those caps) grounded at the same point on the busbar as the caps?

If no (especially if they were grounded to the chassis), you effectively had a ground loop, which was corrected/shortened once the filter caps had different grounds.

[foghornleghorn]

HBP, no, the other preamp grounds went directly to the chassis point. However, the bussbar itself is directly grounded to the same chassis point.

I'll take out the KT66 screen gridstopper and see what that does.

Here's what I'm thinking of in a 2xKT88 version, see attached schematic.

[DummyLoad]

consider use of the correct OT for SE w/ best tap point for UL; 40%-43%. hammond 1627SE  will work.

--pete

[HotBluePlates]

... the other preamp grounds went directly to the chassis point. However, the bussbar itself is directly grounded to the same chassis point. ...

Sometimes, ground is ground. Other times, ground is not ground.

"Best practice" would be to run a filter cap's - lead directly over the the "ground side" of the cathode resistor for the stage(s) that cap feeds.

Sometimes if you ground caps to a bussbar, which then connects to the chassis, but the stages' grounds connect to the chassis somewhere else... well you might get interaction between the ground currents for different stages. This is at its worst when the large, rough charging currents for plate and screen filters modulate the ground current of low-level stages. You wind up getting buzzy hum amplified at the early stages of the amp.

It's hard to predict exactly when a wiring scheme will cause one of these problems. The best you can do is try to minimize the likelihood of occurrance, which is how people wind up obsessing over star grounding and "galactic grounding", etc. To me, something like galactic grounding (as explained by O'Connor or Merlin/Valve Wizard) makes the most sense.

[jazbo8]

consider use of the correct OT for SE w/ best tap point for UL; 40%-43%. hammond 1627SE  will work.

--pete

Then it is ghetto no more  so now we know a cheap PP OPT could be used for non-optimized UL duty, I think that's cool.

[foghornleghorn]

Plus the Marshall OT is in stock, so I don't need to shell out $150 for a 1627SE.

Thanx again for all responses, I knew about ground loops but I didn't know how unpredictable they can be. Another lesson learned!

[HotBluePlates]

... Here's what I'm thinking of in a 2xKT88 version ...

consider use of the correct OT for SE w/ best tap point for UL; 40%-43%. hammond 1627SE  will work.

Then it is ghetto no more  so now we know a cheap PP OPT could be used for non-optimized UL duty, I think that's cool.

Plus the Marshall OT is in stock, so I don't need to shell out $150 for a 1627SE. ...

But you're going from basically 1x 6L6 (at how much idle current?) to a pair of KT88's (at how much more idle current?).

Try it and see what happens. Bass and powe output may be constricted, because you're gonna cram a LOT more unbalanced d.c. through an OT not designed for it. Big SE OT's are BIG lumps of iron... A Hammond guitar-style OT rated for 100w is 4.5 pounds; the 1627SE rated at 30w is 11 pounds.

You don't even want to think about the big-boy SE OT's Hammond has, like the 1642SE rated at 75w and weighing 28 pounds.

[foghornleghorn]

The KT66 draws about 85mA quiescent and not a lot more full out.  I think two KT88s at the same voltage would come in around 200mA or so, which any 100W OT designed for four EL34s in a Marshall circuit should handle OK. (Naturally that would exclude a Twin OT as that isn't really a 100W animal and is physically smaller). Then there is of course the DC at about 350V on top of that, so that combination is indeed the big imponderable!

The Marshall OT I spec'd weighs about 7 lb, and I also have an Edcor 1.7K 100W PP OT that weighs almost 10 lb. I'd rather use the Marshall as I think it's probably tonally better suited to the circuit, plus I have the Edcor earmarked for a 150W Steel String Singer, assuming I can ever get around to it.

[foghornleghorn]

Changed the KT66 screen resistor. Was a 5K1 3W metal oxide. Is now a 360R 2W carbon comp, and sounds better.... more open, somehow more "harmonic", but still crunches nicely when pushed.

[jazbo8]

We need soundclips, man! 

[foghornleghorn]

Sorry for the delay, I now have five soundclips but I can't upload mp3s here. Please go to www.ampgarage.com Technical Discussion, "And Now for Something Completely Different..", they're all there.

[tubenit]

Congrats on the amp tone!  Very very cool sound.  Loved it!   

Thanks for sharing your journey and your success.

Guys, the clips are very much worth listening to, IMO.

With respect, Tubenit

[foghornleghorn]

Thank you Mr. T. My friend Chris Rolin, who did all the playing, is a very talented musician. All the clips were done on his '89 parts thinline Tele with Bill Lawrence pickups except the Benedetto w/Backing Track, which was a Benedetto  Bravo archtop. I made notes on settings etc. if any are interested in those details.

[jazbo8]

Very nice! My favorite track is Blooz, what are the settings? The amp is almost noiseless, very good for a SE build! Love to see some pictures so I can learn something.

[foghornleghorn]

Thanx jazbo8! Settings for all tracks except Benedetto were: amp gain 2:30PM, MV (Trim) 09:00AM. Benedetto was gain 11:45AM, MV (Trim) 09:30AM.

Amp TS settings, all tracks, bass 8, mids and treble both 4 1/2.

OT sec 3K @ 8 ohms into an 8 ohm G12-80 open back 1x12 into an SM-57 mike into an LA-610 Mk II, slight compression @ 3 on the pot.

Here's the current 10-25-13 layout and schem.

Thinking of separating the mid pot from the bottom of the TS and putting it on the other side of the relay to switch it in and out. So then I'd have available a footswitched 6dB gain boost from the switched 1uF cathode cap that's already on the relay, AND a mid boost from the TS, both together. Opinions?

[foghornleghorn]

I did the separation of the mid pot from the tone stack this afternoon. It works nicely, perhaps a bit lower mids than I anticipated, maybe that is the V2 boost which is interlinked with it.

The bass pot now seems to cut the gain quite a bit when turned down, or conversely brings up the volume when turned up. I'm wondering if it should be a higher value than 25K now that it's out there by itself. Any opinions?

It's actually a 50K with a 51K across it, and it comes up quickly and the last third of the turn doesn't do much.