Hiwatt layout assistance: humming and oscillations after rework

[cdemike]

I recently redid my 50w SAP clone to lower filtering to roughly match equivalent power section values of a DR103 and to use the normally unused triode as an EP2 preamp approximation to feed into the front end. During the process I thought I was being clever switching the overall filtering scheme to have individual capacitors for each gain stage in an effort to approximate a star ground.

Unfortunately the amp now has a terrible intractable hum when plugged into the “boosted” input and now has what I’m thinking is an ultrasonic oscillation based on some intermittent artifacts that sound like static. They only appear once the master volume is up over halfway but occur unpredictably in terms of impacting transients and decay. The lack of correlation with a transient or decay and that it’s worse with the master volume up make me think oscillation over a bias issue, and I have reflowed all the solder joints and wasn’t able to get any similar crackles or pops chop sticking the joints.

I’ve reshuffled the preamp tubes which led to some improvement, but I spent most of today trying to get rid of the oscillation without success, including shielded wire on any signal lead connecting to the front panel to a tube socket, 22k grid stopper on V3, a 50pf cap between the PI plates, a 250pf cap across the V2 plate resistor, a 250pf “enhance”cap, and moving the output section grid stoppers as close as possible to the socket. I did get some improvement in stability though these and re-routing some of the connections between the preamp board and tube sockets, but I can’t eliminate the issue and the amp is audibly unstable.

I got frustrated and decided to shift gears and to try to tackle the hum with a variety of different grounding schemes, including a full star arrangement connecting the offending first stage (the EP2 boost stage) both to the signal ground and to the reservoir capacitor cathode. No success here either.

Unfortunately with these efforts at fixing both the hum and the oscillation falling flat, I think I’ll likely need to redo the board to basically be a slight alteration of the original layout. All of the layouts I can find are for the version with the third gain stage sharing a B+ node with the first gain stage, but after some experimenting before this failed overhaul I found I prefer the late 60s version with the 8-10uF cap in a node parallel to that feeding the second gain stage.

Does anyone have a layout for the late 60s Hiwatt preamp board? I’m also all ears if there’s something I might be able to try before starting over again.

[cdemike]

Forgot to post this recording of the suspected oscillation: https://soundcloud.com/mike-913267560/s ... scillation

I did recall that Ceriatone has a complete layout available, though it features the late 70s style PI which I believe is possibly anachronistic (maybe Mark Huss or another Hiwatt expert might be able to weigh in on whether this PI appeared in amps produced in the late 60s?):


I'm seeing enough spare room in their layout to shift things toward the PT side of the board to clear the requisite 3 turret pairs for the cathode resistor/cathode bypass capacitor, plate load resistor, and coupling cap for the EP2 preamp. Assuming I keep the late 60s B+ rail with the parallel V2 and V3 nodes splitting from the PI node (versus the early 70s style where V3 and V1 share a B+ node), it seems like I could just derive a new node for the EP2 preamp by adding another RC pair in series downstream from the V1 node as in my current setup. My hesitation is that I have tried a lot that seems like it should have helped without success, so I feel like I might be missing something.

The schematic as it stands now is attached.


Edited: schematic attachment didn't upload the first time.

[martin manning]

I'm hearing 120 Hz hum, HV power supply ripple.

[cdemike]

Thanks! Maybe a silly question, but I can't figure out why the "boost" node would have so much 120hz hum when the next node "upstream" is quiet. I swapped filter caps at the node to see if I just had a faulty cap yesterday but didn't get any improvement. There's also continuity (<0.7ohm) both between that cap's anode and the plate resistor and between its cathode and ground. What would cause this node to pick up so much 120hz ripple?

Maybe another silly question, but I'm gathering that the crackling noise doesn't sound like an oscillation to you. Are you thinking this is chiefly B+ ripple?

[martin manning]

Yes. Could be a poor or missing ground.

[jelle]

Time to break out the sine wave generator and scope. If you don't have those, imo this is a good time to get them. Issues always pop up, these tools will help you to see what's up and where.

Troubleshooting an amp over the internet is not always easy, but with these tools it becomes a breeze. Facebook marketplace, craigslist, ebay are good places to pick up used ones. I like the older Tektronix oscilloscopes, make sure they work before you buy, and read up how to use it safely. You got this!