2C34 Project taking off...

[tri-comp]

Hi,

I was talking about an upcomming 2C34 Push-Pull amplifier-project here:
https://www.vintage-radio.net/forum/...d.php?p=393266
In the meantime there has been some development.
The chassis was cut, bended and painted as the following pictures will show.
I don't remember ever having seen pictures from the laser-cutting process or even having read a discription on how laser-cutting is done.
Probably you wont find it very interesting without pictures..... which is why I took a lot
The mechanical part is probably 80% or more but it's all done so not in a very distant future I will start actually doing some serious wiring. It will be rather uncertain just HOW the amplifier will look in the end. 2C34 requires some heavy driver-stuff which I have. I'll be using JAN 5687WA, a tough little cookie. Just don't know if it's tough enough. In order to get a reasonable output the 2C34 grid should probably be taken to +20V and 10mA which of course won't happen with that driver. DC-coupled cathode-follower ....or a pair of FET's between the 5687 and the 2C34 ....or something else.
Time for experimenting. As it is my first tube-amplifier it won't be easy.

The photos show the 2C34, 5687WA and Tung-Sol EF806SG

Now the pix.

[tri-comp]

more pictures...

[tri-comp]

...and more

[tri-comp]

final part

[cmjones01]

That's very pretty - beautiful work. I hope you don't mind me asking, but have you prototyped the electronic design before doing all this detailed mechanical work?

[tri-comp]

He, he ... This is the chassis prototype and the electronics wasn't built beforehand.
This will be my test-bed and it will just HAVE to turn into a success won't it?
If it turns-out something is very wrong with design, I do have the opportunity to go and desing and make another.
But with the room inside this chassis and the tube-sockets chosen along with the transformer cut-out dimensions I should be able to get something usefull from this design. Different VVT iron will fit if necessary.


All is drawn in Solidworks CAD design software by my brother who is also the brain behind the laser-cutting.
He knows nothing about electronics which is where I step-in.

So, this is the very first design we made into production.
A much more elaborate chassis was designed before this one and will also be made into a proto-type soon. (EL34 PP with Lundahl iron)
Just, work seems to be more and more getting in the way of pleasure

Btw., the iron used in this amp. is from VVT Ltd.
They are at: http://www.vvttransformers.co.uk/index.htm
I had to take the transformers apart and paint the bells and the core.
Nice transformers at a reasonable prices and sloppy finish compaired with Edcor and Lundahl.
Still with the prices there is room to buy a little Hammerite

rgds,

/tri-comp

[cmjones01]

Well, I wish you the very best of luck, and I hope it works eventually. Just remember that the components have no idea what you are trying to achieve, and they won't necessarily co-operate ;-)

Chris

[Peter.N.]

Looks very nice, we had to cut and drill all our chassis by hand and then modify them when we decided to change the design - I wouldn't know how to use one of those things anyway.

Peter

[tri-comp]

Originally Posted by cmjones01

Just remember that the components have no idea what you are trying to achieve, and they won't necessarily co-operate

Aah, but I have almost 40 years of experience in tricking them into doing what I want

OK, if I have to settle with 5-6Watt output with a normal long-tailed pair as a driver I would also call that a success. It's just that I believe the 2C34 has a lot more potential to be exploited so before I call it quits I'll at least put up a fight

Unfortunately there's not a lot to be found on the WEB about 2C34 projects.

rgds,

/tri-comp

[glowinganode]

That's lovely job you've made.
What value is the transformer primary A-A load?
I assume you're running them in push-pull (AB?) rather than class A strapped anodes.
Rob.

[tri-comp]

Hi,

More pictures as a little was added today.

I really want to run them as P-P, I'm not a SE-person.
Take a look at the attached data sheets for those interested.

OPTransformer data is here:
http://www.vvttransformers.co.uk/vtp..._datasheet.htm
I was figuring that with all these secondaries and UL-taps SOME mix should fit the 2C34 setup.

rgds,

/tri-comp

[tri-comp]

Some more added under the hood.

rgds,

/tri-comp

[tri-comp]

It became time to start messing with electronics.
Since I'm no constructor and have very little experience in designing electronic circuits (I'm used to repair electronics, not to build it) I'll have to ask for your opinions on what I've drawn-up.
The 2C34 should be run in Class-B to get something usefull from it.
That means pulling up to 10mA of G1-current and supplying around 100Vpp G1-to-G1
Since the G1 Bias should be around -15V I suppose that would mean swinging each grid from -65V to +35V PEAK corresponding to approx. -45V to +25V RMS
(Above values taken from the Raytheon data-sheet.)
To this end the Catode-Follower (CF) with a CCS is used, ref the schematics.
Now my question; what necessary current should I expect to be needed through the CCS?
The transistor (MJE340) should regulate easily down to 1mA but let's make it 2mA just in case. The diode string current will be 1/10 of that.
The 10mA of G1-current will be supplied through the CF from B+ when swinging the grid positive but should the CCS be running this current plus the 2mA?
The CF (5687WA) will start sweathing if loaded with a constant-current of 12mA I think, not mentioning the power-transformer that will have to supply 4x12mA extra, one for each CF.

Also, how much negative voltage will be needed for the CCS? Entry point of negative supply is common point of R3/R4/R13 & Diode-strings. Will 125-150V be enough? ..and why?
B+ is aroung 330V. A bit high but hopefully nothing to worry about. I'll possibly have to drop it a little for the CF, depending on the choice of CF-valve. So far: 5687WA.

The component values in the schematic hasn't been calculated yet. When they are I'll post a new schematic with every detail added, this is just a draft.

Phase-splitter will be a concertina-variant with a pentode in front (6AN8A)
Mostly because I ran out of tube-sockets

rgds,

/tri-comp

[glowinganode]

Hi, this is as far as I got with mine.
I decided transformer coupling the grids was a much neater method instead.
This saves a seperate negative supply, several valves and associated components and avoids variation in standing current due to drift in the cathode follower.
Good luck with the design, sounds like you've put a lot of though into it.
A few mA's standing current should be fine.
Rob.

[tri-comp]

Hi Rob,

I remember your project went dead about here.
I do agree about the transformer-solution it's just not so easy to experiment with transformers. You've got to live with what you have. Not so with CF's and CCS. There is a lot of room for experiments. Different valves, CCS with a pentode (...aah, but not enough socket-holes punched out ) and more.
Besides, who can say exactly in the first go, what transformer is going to work in this set-up. Not a lot of experience to be tapped into on the web.

Stay tuned, I'll at least end up with 5 Watt

Best,

/tri-comp

[dominicbeesley]

I was thinking about this earlier this week the cathode follower will need to source a lot of current to pull the output valves grids high but will it need to sink as much - the grid current will be battling to pull them back down?

My amplifier is stuck due to my ropey metalwork skills - I tried to make some cap holders but they broke after about ten minutes!

D

[glowinganode]

No, you're right. This is an ideal application for a cathode follower.
The amount of pull down current is very small, but needs to be high enough to achieve a decent slew rate on the negative going cycle.
It has to overcome the capacitance of the grid - cathode of the output tube and cathode - heater capacitance of the c/f.
A couple of mA's is usually sufficient.
A pentode might make a more linear c/f due to its constant current nature.
Rob.

[tri-comp]

D, My cap-holders would equally break after a few minutes if I didn't have access to a tube-laser. Now feel free to drive over them with a truck

Glowinganode, thanks it is this info about the pull-down current I was searching.
The semiconductor CCS i drew-up is just a draft.
Morgan Jones knows just how to make them i real life, ref attached doc.

PS.
Don't really know if publishing this info on the forum is OK with copyrights.
If not, moderator feel free to delete.

rgds,

/tri-comp

[glowinganode]

I found this quite interesting.
Rob.

[tri-comp]

I already read these interesting pages.
Nothing beats solid experience coupled with deep insight.
Still, the PowerDrive may seem more like an OverDrive to my poor 2C34'
They are already split down the middle right to the common katode

I can't locate the 10M45 CCS-chip locally unfortunately. Otherwise I would use it for sure. I've already got some Texas Instruments TL783 but they are only good to 125V, at bit on the low side.

rgds,

/tri-comp