audio-talk

So this is the new thread that branches from my 6C33C build project; taking a lead from Steve C's breadboard 13E1 project I'm going to go down the same route (but leave the door open on the 6C33C project).

So here's the starting point - today I received a nice pair of 13E1 valves.

The 13E1s look rather splendid in the chassis that I had machined for the 6C33Cs;

Some of the other ingredients;

• Driver stage uses a Broskie Aikido topology with 6SN7 valves. Power supply for the driver stage will be a Neurochrome Maida regulator.

• 13E1 B+ supply will be CLCLC, basically re-using the power supply configuration of my Transcendent 300B SE-OTL amps (that have similar voltage/current requirements).

• 6SN7 filaments will be powered with regulated DC. 13E1 filaments will use AC.

• Power transformers will be located in a separate chassis.

Anyway, lots to do!

Ray P wrote: ↑Sun Dec 09, 2018 6:44 pm Some of the other ingredients;

• Driver stage uses a Broskie Aikido topology with 6SN7 valves. Power supply for the driver stage will be a Neurochrome Maida regulator.

• 13E1 B+ supply will be CLCLC, basically re-using the power supply configuration of my Transcendent 300B SE-OTL amps (that have similar voltage/current requirements).

• 6SN7 filaments will be powered with regulated DC. 13E1 filaments will use AC.

• Power transformers will be located in a separate chassis.

Anyway, lots to do!

Hi Ray,
I would try CLC in the first instance; basically half the components of the Transcendent power supply. I say this because I tried the CLCLC yesterday and the sound became, for want of a better description, somewhat "slugged" The whole presentation with stereo 13E1s is much faster and more dynamic when just the CLC is used.
My own power supply is basically half of the mini-beast, same value caps and the same Triad Magnetics choke.

Thanks for the insight Steve, something to think about and consider as the project progresses.

I actually have the CLCLC power supply parts already and mostly assembled so I'll go with that to start with. The 300B SE-OTL and minbeast power supplies are different so I'll put off buying new chokes until I get a flavour of what the 13E1 has to offer.



Just read your update on your own fettling; interesting that lowering the dissipation had no discernable affect on the sound so maybe there is a reasonable sized envelope in which the amp performs? As my B+ might be a tad higher than yours I was planning on using 2R7 resistors for the current setting, at least to start with, to keep the dissipation a little conservative.

I've noticed that the bottoms of the 13E1s are much less profiled than the 6C33Cs and that means that, whilst they plug snugly into the sockets, the glass envelope is in contact with the chassis at the seven bolt locations. I'm wondering if this might result in some localised thermal stress to the envelope - any thoughts.



There's an easy remedy - just move the sockets to the top of the chassis, though that is less aesthetically pleasing.

I would have thought if you have sufficient pin contact with the tube fractionally raised above the chassis, some spacing fibre washers could be used on the pins or for convenience, superglued over the pin holes of the base.

Or can you mount the sockets on top?

Could have the valves sat on a thin silicone o ring that is large enough round to sit over the bolt heads?
The red ones that folks use as tube dampers

I agree it's a bit of a worry and Greg's advice seems like a good option. I went looking for Teflon ( which I think is good to 250°C ) and came across this :
https://www.ebay.co.uk/itm/Bespoke-Cust ... :rk:2:pf:0

I think I'll do the obvious thing and move the sockets to the top of the chassis. It won't look as nice but the sockets themselves have a nice finish and are fairly thin so it won't be too bad. I'll have a look tomorrow evening.

So, back to practical stuff, I have two functioning filament supplies for the 6SN7 valves. I'm using a couple of TPS7A4700 LDO Low Noise Power Supply modules that I've had hanging around, probably a little overkill but they needed to be used. Steady at 6.34V - 6SN7 filaments draw 600mA but after an hour or so the supplies are only a little warm so it doesn't look as though they're being stressed - they're capable of upto 1A - heatsinking is via an aluminium block sandwiched between the PCB and the black aluminium plate.

I've assembled the Neurochrome regulator PCBs this morning. Just need to drill the metalwork that they'll be bolted to (and will act as the heatsink) and I'll be able to test it...

Be nice to see this one moving.