Hiwatt Dr103 clone oscillating

[bxershrts]

Hello everyone,

I built a a Dr103 clone and it all works and sounds great except for the master volume oscillating at max.
What mitigates the oscillating when the mv is maxed is with the treble at max as well.
If I have the mv all the way clockwise and the treble all the way clockwise the amp functions great, albeit bright and lacking controls autonomy. When I turn the treble counterclockwise the oscillating starts, then at which point if I turn the mv back slightly the oscillating goes away. I built it in the same way as the originals and used vintage Mullard mustard coupling caps, perhaps one is leaking DC.. - any other thoughts on things to check, thanks again!

[pompeiisneaks]

Have you tried swapping the OT leads to make sure it's not a phase issue? Not sure if there's NFB or not on the amp.

~Phil

[bxershrts]

Thanks for your reply Phil if you can view my photo, I kept the blue wires long at the board until I knew they were oriented correctly. That noise is unmistakable

[Oddvar R]

Very nice looking build. I don't know how you do that...

[CraigGa]

Very nice looking build. I don't know how you do that...

My thoughts too, that's very neat looking build.
It's a lot like the layout of my Burman, I can see that they may have peeked inside a Hiwatt at some point

Craig

[Stevem]

If swapping the output wires from the PI does not kill the issues then you need to pin down if its stemming all in the preamp section, or from the output section impressing a signal into the preamp section.

Things can be very impedance sensitive, so if you yank out two output tubes and what you have reported about changes a bit then you have the preamp and output section working against each other.

Long PI output wires are a sure fire invitation to oscillating problems!

Also in terms of a original Highwatt build if you followed that, the daisy chaining of the grounds is a very bad thing Also!

[mhuss]

That is a nice looking build (except for the blue wires )

I have never had an original Hiwatt or a clone that followed the original grounding scheme have an oscillation problem like you describe; I've only seen this when an extra gain stage or an active effects loop was added. Did you follow a layout or...?

In the original preamp layout, the circuit ground is connected to chassis ground near the input jacks, and daisy-chained from there across the board. All grounds, including the pots, connect to that ground chain based on their position in the circuit, i.e., each gain stages' grounds are close together. The preamp dual filter cap is also connected to that single chassis ground.

It looks like you have the power grounds connected as per the original.

[bxershrts]

Thanks for all your replies. Yes I kept the wires long until first fire up to determine the orientation. Here is a picture of it now.

My first instinct is a bad wiper on the mv pot as I have run into that sound before. There are a lot of pictures online of the Dr103. I use it all when i build, pictures, schematic and layout.

[bxershrts]

I just want to be clear - the amp only oscillates at maximum on the master volume, turn it back 10 percent and its gone. If I turn the treble all the way clockwise with the master all the way clockwise there is no oscillating.

With the plate leads oriented wrong the amp will oscillate instantly. It is not that.

This was a fun and epic undertaking. I have built the whole amp. From chassis to circuit boards and faceplates down to powder coating and cabinet and tolexing. I had most of the parts in the shop already and did a lot of reading on transformer choice. The amp sounds wonderful just that 5 percent is missing on that darn master volume.

[pdf64]

A good approach is to identify which is the last stage, and which stages are within, the unstable positive feedback loop. Then analyse the schematic and layout to identify where the excessive degree of unintended coupling might be occurring. Then use chopstick and a small piece of insulated screening sheet metal (connected to the chassis by a suitably long wire) to identify the critical point/s where the actual loop is coupling. Or temporarily clip / tack in small caps to slug the response.

Dunno if it’s relevant but modern Russian 12AX7 can have rather high gain; perhaps higher than typical BVA the amps were designed around? Cascade a 5 or 10% gain increase through 2 or 3 stages, and if the original layout was only marginally stable, just that may be enough to push things over the edge.

[Stevem]

Am I reading those grid resistors right on pin 5 of the output tubes, there 21.5k?

[bxershrts]

Steven - yes they are. Mil spec. On measuring an original they all came in at under 22k - i ran out of 22k resistors and used 1 watt 21.5k Mil spec, measuring the same as the original.

[pompeiisneaks]

BTW if you look at the note by mhuss earlier, he's the person that wrote the outstanding hiwatt.org site and colleceted great detail on the Hiwatt amps.

https://hiwatt.org/tech2.html

that's the link to the known good layout he's referring to. If you followed that to a T, then I'd guess it's some component that's bad in some way, causing the problem.

Have you chopsticked components or leads to see if any cause the issue to get better or worse?

https://hiwatt.org/Layouts/DR103hiwattlayout.pdf

Seems similar to yours, but I've not compared it 1 to 1, just saw many similarities.

~Phil

[pdf64]

So the presence control setting doesn’t affect the issue?
If so, that kinda indicates that the power amp isn’t in the loop.

[pdf64]

... Long PI output wires are a sure fire invitation to oscillating problems! ...

If they’re twisted (or otherwise in close proximity with each other) and well away from preceding stages, I don’t see that their length per se should be an issue?