• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

304tl se amp

See here:
http://www.alumrocktech.com/amplifier.shtml

1a.jpeg


schematic-1bit-20020701.gif
 
I have two separate PSUs and they are stacked to avoid the large biasing resistor for the 304TL (just think to the Loftin White and variants). The EXO99 choke and that Rx are fixing the bias for the 304TL, to make it more clear

801A grid:-9V
801A cathode: 0V
801A anode=304TL grid: +280V
304TL cathode: +330V
304TL anode: +1200V

The OPT is returned to the 304TL cathode.

I loved that amp but I am not using any more 304TL as they were way too hot :(

In the same amp I used GM70, 813, 304TL and 75TL. Going to give 845 a try soon.

Disclaimer: the high voltages can kill! Beware!

Gianluca
 
Gluca said:


Good luck with the OPT, not easy to find a good one.

Gianluca

That is a bit of an understatement. Either massive idle current, or massive B+ needed to reach anode-glowing dissipation. Insulation for multi-kV B+ is not good for leakage inductance, and neither is the size of the required cores for the SE power this one id capable of.

Neat thing to consider is only lighting half of the anodes and making a 152TL from it to start with. Slightly increased Miller from the unheated grids...and do use something bigger than a 6SN7 to drive it.
cheers,
Douglas
 
That AlumRock amp looks attractive in a kinda commercial Bell & Howell way. The performance in output power is terrible however and this has to be due to insufficient drive and B+. This amp shud be making an easy 100 watts RMS/ch. I also do not like the total lack of attention in the way the IR from the output tubes will bake the black OPT's and other nearby surrounding components. Heat is the enemy to electronic components and many materials. A simple mirror finish SS heat shield would do wonders here.

I think the most practical answer to the high voltage on the OPT issue with these electric chair SET amplifiers is to parafeed the output stage and get the HV completely off the xfmer with a quality capacitor.
 
rcavictim said:
I think the most practical answer to the high voltage on the OPT issue with these electric chair SET amplifiers is to parafeed the output stage and get the HV completely off the xfmer with a quality capacitor.

A parafeed transfomer can be materially smaller (than a series feed one) and exhibits a larger inductance: lower and upper end gets better.

I am with you about heat: I am not using 304TLs for this very reason. Power is dissipated to bias the tube, to light it and to have a decent DC supply.

Gianluca
 
GaryW said:
Nice. 6K primary impedance seems lowish - Rp is on the order of 1.5K depending on the operating point. Also, a hum pot on a 10V 12.5A AC filament seems optimistic, as does just an input tranformer for gain and driver.

hey-Hey!!!,
The 304TL is 2A3-ish in plate Z. 3x the gain and gm leaves R_p the same. Running them with filaments in series is asking for it; two triodes go ~7.1 volts positive, and two go 7.1V negative. Far better IMO to have the four halves of each filaments going half that amount. Run them series with DC and two plates will cherry nicely whilst the others are dark. Parallel filament operation side-steps these headaches.
cheers,
Douglas
 
GaryW said:
Nice. 6K primary impedance seems lowish - Rp is on the order of 1.5K depending on the operating point. Also, a hum pot on a 10V 12.5A AC filament seems optimistic, as does just an input tranformer for gain and driver.


Hi GaryW, just to let you know the schematic tomtt posted is a Sukuma design. He's ideas are held in pretty high regard and they seem to work very well.


We all can have are own opinion I'm just saying you may open a can of worm with some other people.


I would agree though a hum pot seem fairly ridiculous in this application but it would be worth a try. And if it didn't work your only out a few buck's verses building a whole dc supply.



Also 6K plate to plate load will probably be a bit high even if the tubes are biased in class a. Just looking at the data.


Nick
 
nhuwar said:



Also 6K plate to plate load will probably be a bit high even if the tubes are biased in class a. Just looking at the data.


Nick

I think a review of your methods is in order. 6k a-a leaves a 3k load for each tube in class A. 900 vac with a cap input filter would give us B+ ~1k3 V. At 300W plate dissipation we'd be allowed an idle current just over 200 mA. V=i*R and that would take us down to ~750 V at maximum conduction...That would leave it AB. So kick the a-a load up a bit to stay Class A with that idle point.

One thing Sakuma-san was noted for was amps of restricted/limited bandwidth. Given his likely music taste it was of little issue I suspect...just not what would make it worth my time for such a project.
cheers,
Douglas