Shiva/Hot Cat 30 build

[Zozobra]

A year and a half ago I started building a Shiva pre/Hot Cat 30 power amp and then when I was close to finishing it I had to move and then didn't have a workshop to tinker in
Well that changed recently and I now have a workshop again so I've been finishing up the things that I had to abandon and this was high up on the list. I've stuck all the pictures and information in an Imgur album as that seems to best way to document these things:

Shiva/Hot Cat 30

I fired it up yesterday and it worked fine but the voltages are a bit too high for a 3k25 primary OT as I was seeing over 400VDC on the plates and screens. My first thought it to try and shave 30V off by putting a string of 3 5W 10V Zener diodes between the -ve and ground of the bridge rectifier. The OT has a 4k primary option too so I think I'll switch to this to avoid over dissipation. I think that I should be safe with 15W of Zener dissipation to drop that voltage. The PT is rated for 370 mA so if I was to drop 30V at that current the dissipation would be 11.1W so I have a bit of a margin. I can't see the amp pulling anywhere near that much so I guess I should be okay. If I'm waaay off the mark here could someone please let me know

[DRH1958]

I see in your notes at the end in the Imgur file that you are considering adding a 220R for sag. Why not install that and see what voltages you end up with? This just might get you where you want to be.

Also, I think you would want to have additional nodes to your PS for decoupling the preamp stages.

And also it says you aren't satisfied with the MV. If you add that cap in series, it turns into a treble cut similar to a Vox AC 30. It should just be only the pot for the crossline MV.

It might be best if you have the whole schematic drawn. I don't see anything including the power tubes or the connections from the PS to all the different sections of the amp.

EDIT: Now that I look closer at the actual build, are those black caps and resistors at the top of the preamp board other nodes for the preamp? Didn't see them at first.

[Zozobra]

It's basically a race between the zeners and a 220R resistor in the mail at this point as to what gets done first

As you spotted in your edit each preamp tube has its own decoupling node.

I'm not sure I follow your comment about the MV. My plan is to add a series resistance to the hot lug of the MV pot to try and give it more usable range. Will that not work? This is for the pre PI MV btw.

Thanks for you comments and observations I'll draw up a complete schematic when I get the chance.

[DRH1958]

Yes, for the pre PI MV, you just use the pot wired as you have it, a variable resistor but eliminate the cap. That 4N7 cap basically turns the MV into a treble cut control. Vox uses this on some or most of their AC 30s. It is labelled "cut". That is on your second page correct, drawn by hand? You have it wired in just after the 220k grid resistors and 10k grid stoppers. It should have all the range you need if you eliminate the cap. It goes from open for full vol. to eventually shorting the two outputs together for zero volume. Everything in between is different volumes.

[Zozobra]

Ah yes, I see what you mean now. Yes, I am using that a cut control although it doesn't appear to do much over the range of the pot so I may experiment with a few cap values for it.

My intention is to add a resistor between the treble pot wiper and the hot lug of the volume 2 pot on the preamp schematic to reduce the signal fed to the PI.

I'll draw up a full schematic tomorrow to avoid the confusion of working across different sheets.

[DRH1958]

Sorry, now that I re-read my post, I meant to sat "POST" PI MV in my first sentence of post #4. I thought you mistakenly added that 4N7 cap to it and turned the crossline MV into a cut control. I thought that vol following the tone stack was a second pre vol since it was labelled vol 2 and not MV. Yes, it will be great to have it all on one sheet.

[pdf64]

Regarding the pre phase splitter / type 4 master volume's poor resolution, have you tried a 1M 10% audio taper? Check resistance at 50% mechanical rotation.
Poor resolution at low settings may indicate a higher %, even linear taper; or just a crappy pot.
CTS and Bourn standard is 10%, Alpha USA is 15%.

[Zozobra]

I'll check the pot taper tonight - It is an alpha 1M log taper iirc. I'm sure I checked it to make sure it wasn't junk when I fitted it. My Zeners and sag resistor have arrived in the mail so I hope to have everything put together this weekend.

[Dave H]

Yes, I am using that a cut control although it doesn't appear to do much over the range of the pot

Do you have it wired for max treble when fully ccw like a VOX? I think the range is better (with a log pot) if it's wired for max treble when fully cw (connect to the wiper and ccw lugs) and it's less confusing too.

Also I think it needs a capacitor between the grid of the bottom PI triode and ground.

[Zozobra]

I'll give that a go, thanks.

Yep, there is a 100n cap between grid and ground of the 2nd PI triode. It's not shown on the schematic as I was originally going to have a second channel feed like the hot cat 30 but then didn't bother.

[Zozobra]

Also at some point once I'm happy with that everything is running properly I will adjust the preamp to taste. One thing that I have been wondering about since I first saw it is the use of caps to the ground part of a potential divider as seen the Splawn Nitro as shown here:



Does that shunt treble to ground in a similar way to how a treble bypass cap works in the upper part of the divider? If it works that way it looks like a nifty way to increase bass response without butting up against blocking distortion in the preamp.

[Zozobra]

Just so I might actually be able to use this at home for more than 3 minutes at a time I'm thinking that I might make the cathode resistor network switchable so that I can use 6V6s. If I use the 4k primary and double the speaker load then I can get an 8k primary which should be okay with 360-370VDC on the plates and screens with appropriate biasing. I'm currently using a 360R bypassed resistor for each cathode when using EL34s. If I connect the cathodes of both 6V6s to just one of them so they share 360R then that should be close enough, right? Also should I give any special consideration to the switch I use?

[pdf64]

I think that 360 ohms per 6V6 may be ok for cathode bias? If not, there's nothing too nasty about the cathode circuit and the current / voltage is lowish, so almost any quick make / break type suitable for dc use should be fine, eg even mini toggle types.

The 1nF V3 grid-to-0V cap will roll off high frequencies, which may otherwise cause harsh distortion at the clipping stage.
I don't think it's boosting the bass response, except in a roundabout way.

[Dave H]

One thing that I have been wondering about since I first saw it is the use of caps to the ground part of a potential divider as seen the Splawn Nitro ...Does that shunt treble to ground in a similar way to how a treble bypass cap works in the upper part of the divider? If it works that way it looks like a nifty way to increase bass response.

I think it will do the opposite. The time constant of the top RC is lower than the bottom RC giving a shelving treble boost/bass cut.

[Zozobra]

Okay, thanks. I will experiment
It's still damn loud using 6V6's and the MV range leaves a lot to be desired so I think I'm going to add a divider after the MV to tame the signal fed to the PI.

With a bit of luck I hope to get down to my local Fab-Lab to make a sleeve for this to live in this weekend. Too many of my amps are strewn around my workshop in various states of undress!