up-to-date El Cheapo schematic

[Eli Duttman]

A 6V6 family tube was desired for the "finals". The 12AQ5 was settled on, for its low price. Any 6V6 family tube works with the parts values shown in the schematic.

The B- rail could have been quite a bit "shorter". What was settled on came (you guessed it) from cost considerations. The fact that the B- supply is technically nice is a bonus.

As for "fixed" bias versions of the basic topology, refer to Poindexter's "Musical Machine". Poinz was 1st with this style of 2 stage amp and MUCH was learned from his efforts. The "Musical Machine" is an all out, cost be damned, assault on performance. "El Cheapo" is a budget project that performs far better than the money spent suggests would happen. While inexpensive, the PSU is anything but cheap.

 

[Jim McShane]

All the small components are available from Jim McShane, and the trafos and such can be ordered from online electronics vendors.) Jim's page for the El Cheapo is here:

http://pages.prodigy.net/jimmcshane/elcheapo.htm

Actually my page is www.mcshanedesign.net/elcheapo.htm, the old link Jeff posted is dead.

 

[jdrouin]

How many watts per channel does a stock El Cheapo put out (say, with Edcor transformers)?

 

[vinylkid58]

How many watts per channel does a stock El Cheapo put out (say, with Edcor transformers)?

~Ten, AFAIK.

jeff

 

[Eli Duttman]

How many watts per channel does a stock El Cheapo put out (say, with Edcor transformers)?

Approx. 6 WPC for triode and 12 WPC for UL. Some variation will occur depending on B+ rail voltage and selected O/P "iron".

Dynaclone Z565s are the "Gold Standard" for O/P trafos, in an "El Cheapo". Edcor's CXPP25-MS-8K is certainly hefty enough for UL mode. Stick strictly to triode mode, when low cost guitar amp "iron" is used.

 

[planet10]

Our variation on El Cheapo uses triode wired EL84 and runs in Class A. 3.2W at clipping and a fantastic sounding amp.

Eli, i'll be happy to update the drawing but i can't do it until i resolve some issues with my CAD software.

dave

 

[Eli Duttman]

Dave,

If another Scott "donor" ends up on your doorstep, try 6CM6s as the "finals". The 6CM6 is a Noval based member of the 6V6 family.

 

[tomlang]

Is there any "simplification" that can be done to this? I.e. utilize a LM334 and skip the negative supply for the LTP as I show in post #20?

 

[Eli Duttman]

AFAIK, it will NOT work, without the B- rail.

 

[crazyfrog]

Sorry...

I didn't want to bring back to life an old thread but I had to ask: did someone build an El Cheapo "on steroids" ? With EL84m or 6P14P-ev , higher voltage, fixed bias and MOSFET buffer (ZVN0545A, FQPF2N80...) between LTP and OT ?
If this is the case, is the schematic very different from post 20?
Thanks
Pierre

 

[Eli Duttman]

Pierre, are you "looking" to squeeze 20 WPC out of the 7189 equivalents, as Sherwood did in their S5000? If so, you will need hefty O/P "iron" that can handle upwards of 35 W. The GNFB loop mandates that, in order to avoid core saturation. Edcor's CXPP50-8K is suitable.

 

[crazyfrog]

Frankly my goal isn't the max power but getting a bit over 15WPC shouldn't be too difficult with two 6P14P-ev in push-pull. I'd like to use vintage irons from old amps like PACO SA-40, Pioneer SX-82 or Kenwood KW-70 (real mess under the hood!) I saw recently ( actually these amps are supposed to be fully functionnal at decent price). Maybe I'll make a move but I'm not ready right now to build the El Cheapo.

 

[AquaTarkus]

Pierre, are you "looking" to squeeze 20 WPC out of the 7189 equivalents, as Sherwood did in their S5000? If so, you will need hefty O/P "iron" that can handle upwards of 35 W. The GNFB loop mandates that, in order to avoid core saturation. Edcor's CXPP50-8K is suitable.

I've seen a pair of EL84Ms producing close to that at 420V. IMO hefty OP iron is always to be recommended, for best linearity and frequency response.

 

[crazyfrog]

Eli?

Hi, Me again. Ok, if I put a mosfet buffer after the LTP, what would be the "right current" ? Is CCS better in that case? If a resistor is connected to ground instead, is 470k still a good value for fixed bias?
Thanks
Pierre

 

[ballpencil]

Hi crazyfrog,
That is not the correct way to put a mosfet buffer. It won't bias properly. I can't explain it as i am using my phone. Perhaps look for Class AB2 or A2 schematic for an idea. This is one example A2 Buffer

 

[Eli Duttman]

The easy way is to DC couple the FET to the driving triode's plate and cap. couple the FET to the driven O/P tube. All that's needed here is the impedance transform, as positive grid current is not in the picture.

The tweaked RCA phono setup shows the way.

 

[crazyfrog]

About current and gain

Thanks for your answers. Pete Millett's board has a CCS and Eli uses resistors. Ok for bias. But forgive my ignorance , I don't know what's is the right current for that buffer to do its job for keeping gain at good level (ref. fixed bias).

 

[Eli Duttman]

Look at the ZVN0545A datasheet. Make certain you stay well within the safe area of operation. The drain voltage is known. The gate is at the triode's anode potential. The source will set up a bit below the gate, but not enough to worry about. Size the source load resistor to keep dissipation within safe limits. 1.5 to 2 mA. of drain current usually works well.

 

[crazyfrog]

Oh yes! it's similar to Powerdrive from Tubelab (FET source follower). I used a 2sk3564 at the time to build a 45 SE. My God! I'm not really a tech guy...
Eli, BTW I prefer to use something a bit "beefy" like FQPF2N60C (or equivalent) than ZVN0545A. Since I use 6P14P-EV, the voltage could go over 400V.
Thanks!

 

[crazyfrog]

on second thought, I'll try to find a low current FET like the ZVN...but working at higher voltage.