Princeton Reverb Clone Build Hum Problem

[justsoldern]

OK the plot thickens.

So I have the hum with V1 and V2 removed. Grounding before or after the 3.3 M silences it.

So with V1 and V2 removed I disconnected the heater string going to those tubes at V3. I still have the hum! So I've proven every which way I can think of that it's not the heaters bleeding into anything (At Least I think so).

So I reconnected the heaters. Then with V1 and V2 still removed I disconnected the B+line running to V2 and V3. I disconnected at the spot D on the schematic near the V3 plate resistors (just below the 10 pf cap on the schematic). That killed the hum! So that last B+ run is bleeding into the signal path. How I don't know. Why my speaker output is 60 hz and not 120 hz hum I don't know. I'm thinking maybe replace the cap can next, but it's a $35 experiment. Here is the trace from the speaker output I got with the scope I borrowed which shows 60 Hz.

[loudthud]

Does the heater winding have a center tap? It is usually a green wire with a yellow stripe coming out of the transformer. You should get 3.2VAC (maybe a little more) on each side of the heater wiring.

[Chuck H]

LT probably has the right course. There's something I would check just because it's easy... With V1 and V2 still out, is there less diminishment of hum on either side of the 3.3M? And, what happens if you ground the center lug of the reverb pot?

[justsoldern]

LT probably has the right course. There's something I would check just because it's easy... With V1 and V2 still out, is there less diminishment of hum on either side of the 3.3M? And, what happens if you ground the center lug of the reverb pot?

Chuck H: Grounding either side of the 3.3M and the amp is dead silent. Can't even tell it's powered on. Grounding the Reverb Wiper has no effect on the hum.

Loudthud: yes I have a center tapped heater secondary. I originally used it to ground, but I've tried taping it off and using 2 x 100 ohm resistors and also raising the DC reference to 50 volts with both configurations. I recently added a hum dinger so right at the moment the centertap is taped off and i have one 100 ohm resistor and one trip pot set to about 100 ohm and the junction of those is going to the 50 volt DC offset point. I figured that was my ultimate best shot at it, but I haven't been able to dial out the hum - or effect it at all.
To do the voltage check on the transformer heater output do I need to disconnect the secondary wires and get those voltages with nothing hooked up?

[Chuck H]

Chuck H: Grounding either side of the 3.3M and the amp is dead silent. Can't even tell it's powered on. Grounding the Reverb Wiper has no effect on the hum.

Well that limits the source of hum entry to the grid of the mixer stage and excludes the reverb recovery stage, which is the OTHER triode of V3. And it's 60Hz. This is a corker for sure. I'm currently at a loss, but I'll keep it in mind and be following along.

Could you try this for me? Lift the end of the 470k reverb circuit resistor that joins the reverb pot from the board. Leave that end floating, tack a lead to it and ground it to the input jack ground (no shared grounds) and report on the hum.

Also, have you looked for 60Hz on the HV rail?

[mozz]

How about this. Are you sure pin 4/5 is wired to the next tube pin 4/5. Pin 9 to pin 9, outputs pin 2 to pin 2, 7 to 7.

[justsoldern]

I looked at the heaters pin wiring and it's correct. Also tried the the grounding of the reverb 470 to the input jack. no luck there either Since it quiets when I unhook the B+ to V1 it has to be riding on that wire. My Fluke 115 can't measure any A/C on that line at D but it does at the diamond and square nodes earlier in the B+ circuit. I get 18 mv at the diamond. But the scope trace shows 120 hz. Very confusing

two things next. Maybe use a piece of shielded wire for that last B+ run (grounded at one end only. Or I'm thinking to replace the cap can.



Here is the inside

[justsoldern]

OK so I tried using a shielded wire to run that last leg of the B+ supply from V3 to V1. and that didn't help.

My next step is to get a new cap can. I think I may even pick up a new PT while I'm shopping. Couldn't hurt to have one on hand anyway. I think I may try a Hammond. This one is a Classictone.

Anyone know the best technique to measure low mv level A/C on the B+ line? My Fluke 115 can measure higher levels of it upstream at the nodes nearer the PT using the VAC setting, but for low mV levels riding on the B+ down near V1 the VAC setting reads 0.000 and the mV A/C setting on the meter doesn't seem to work in the presence of the high DC. As if it's overloading. If I do touch the B+ at V1 with the meter set to mv A/C it quiets the amp. It seems like the meter has a low resistance to ground or something.

It could be a week or two before I come back with any updates.

I REALLY want to thank all you folks for the help. You guys are great! I know ideas are getting scarce

[Chuck H]

It's probably not going to help much to replace parts in a build with all new parts now. It's difficult to see much in the photo because of the under board wiring but I'll give it some more time. The first thing I notice is that the many of the eyelets aren't "wet" by the solder. I think you should touchup all the solder contacts with some proper heat and flux. Did you use RoHS compliant solder? (I hope you didn't have to work with that stuff) You need to be sure all the contacts being heated are of a temperature that the solder melts to them. Look at it as heating the parts so they melt the solder, not melting solder onto the parts. That doesn't mean you can't use the wet solder to heat the parts if it's easier (it often is). Just remember that all the parts to be welded must be at solder melting temp before you remove the iron. This is one VERY good reason to use a good, hot iron. I solder plain 60/40 at 710F.

[justsoldern]

Thanks for the input Chuck. I have a nice Weller with adjustable heat. Usually I set it between 700-750 for eyelet work. I use more to solder to a pot or something big And I use a benzomatic torch with a soldering tip for the cap can. (I heat the tip then turn off the gas and work quickly. - rinse and repeat when the tip cools) When I originally assembled the board I used the upside down method like I've read fender did . It leaves those nice round blobs sticking up (which I thought was cool!) But inevitably a lot of the hook ups had to be done with the board in normal position. And now I've played with many of these troubleshooting tricks and resoldered a bunch of things - so it's getting uglier with the solder trying to suck down into the eyelets. I use a solder sucker on the eyelets where I've been constantly redoing the joint - just to pull up as much as I can from under the board. I looked at the solder. It's an old spool I've had for years dated 2006 so maybe pre-ROSS but I think it's good quality - and have used it for years.

There is an under board trace going from the 0.022 to the 3.3M out of V1B I've been thinking maybe somethings not right there. I'm loath to pull the board up to get at it and redo it. I have however soldered in an above board jumper to be sure the connection is good- and of course it didn't help.

Having some extra parts around won't be that big a deal. But I'm really out of ideas, so i'll give it a try. It's a very low level hum and the amp is killer at regular stage volume. I may just have to live with it. I had a lot of fun building this - including the finger joined cabinet and everything - that took some learning. This is one of Leo's best in my opinion

I want to thank you for all of your help.

[justsoldern]

Here's a couple of pics of the non electronic part of the project. It's a really cool amp to have.

[Chuck H]

Well that's just as pretty as it can be

I had a thought... I never met a Fender reverb amp that didn't hum a little (they're not always sure of the words? ). Then on a project of my own I experienced a similar issue. I've often suspected the location of the reverb drive transformer. Not that the power to that transformer isn't filtered, it is, but they still need to "transform" and therefor emit EMF. My issue wasn't bad enough to mess around with it, but I though about removing the mounting screws and moving that transformer around to see if it affected the small hum. Another thing... Because the PR uses a can cap you're forced into grounding the preamp filters with the power amp. It's standard practice in modern design to ground preamp filters independently with the preamp circuits AWAY from the power amp. So there's that. It might be worth trying by connecting remote caps temporarily and ground them with their own circuit on independent leads. I know the hum detected is 60Hz but it can't hurt to try.

[justsoldern]

OK I've had a breakthrough! Adding another stage of filtering to the B+ made it quiet! I disconnected the last B+ wire that runs from the point D that feeds the two 100K plate resistors to V3. Then from there I added a 10K 3 watt resistor and a 10 uf electrolytic cap to ground, then continued on to the point D feeding the two 100K V1 plate resistors. I grounded the 10uf cap to the same spot where the V1 25 uf cathode cap and 1.5 K resistor are tied to ground.

Dead quiet! I can't believe it! I'm not sure I fully understand it. Am I shunting some low level ground loop to ground? Just getting rid of a last touch of supply ripple? (that should be 120 Hz?) Am I shunting some 60 Hz hash to ground that the B+ has picked up? I don't know for sure - but it worked. I need to order some parts to make it permanent and clean. Maybe I'll try another cap can and see if that fixes it without having to add the extra parts. But at least I have a solution and a nice quiet amp now!

Thanks to everybody! This forum is really cool

[Chuck H]

You may not need to ADD a filter, per se. You may just need to ground the actual (original) node D filter where you have the new filter grounded. If the can type filter prevents you from doing that you can just use a different cap that can be individually grounded for that node only. Then you wouldn't have additional resistance in the HV rail to your preamp. Instead of an unused filter node, a resistor and another filter node you would have an unused cap on the can and a remote cap for the preamp filter node. Keeping the circuit schematically correct. It could be worth a try.

[justsoldern]

Thanks Chuck, Good thoughts. So I took out my additional filter. The amp is now stock. No elevated CT for the heaters or anything (just two 100 ohm resistors)

The 4 x 20 uf cap can doesn't let me access the individual grounds. So I disconnected the 18K resistor feeding the last cap in the can near point D coming off the cap can. So now I'm only using 3 of the 4 caps in the can. Then I added back a stand alone 20 uf cap at that spot in the B+ supply. I grounded it near the same grounding spot used for the V1B / V2 cathode grounds. Nice and quiet.

Then I tried it again with the cap grounded near the power transformer. Nice and quiet again!

It seems to be as simple as a bad cap in the cap can! I'm a little dumbfounded by all the symptoms, but definitely happy to have a solution!

I've read older posts here from the gentleman at Mission Amps that he has had problems with bad new cap cans in the past. Count me in.

I also notice on power down that the B+ rail is taking a lot longer to bleed down with the substitute cap than it did with the full cap can in use. I've got the raised DC heater circuit removed for a while now. But even so, the bleed is slower with the substitute cap in there. So I have a new cap can on order, should be here in a few days and hopefully I'll have a nice quiet stock amp. I'm psyched!