Hi everyone 
I'm planning on making a PP 813 amp, triode connected
Using the data here (or attached) : http://www.pmillett.com/file_downloads/813 curves.gif
What sweet spot would you go for (and why) ?
Cheers
I'm planning on making a PP 813 amp, triode connected
Using the data here (or attached) : http://www.pmillett.com/file_downloads/813 curves.gif
What sweet spot would you go for (and why) ?
Cheers
Attachments
Pete got 20W SE triode @ around 900V using a 5k OPT.
The same operating point but in PP would give 40W class A into 10k. Maybe a little more in AB and lower load, but Pete said he thought 5k was already low for SE.
To get more power, you would need more volts but beware the rating for g2. In triode, 1kV would be safe, 1100V is pushing it.
75W triode is very ambitious, maybe in AB2?
If you're very courageous, you can try this one (pentode) Champ 1000 Watt Tube Amp
The same operating point but in PP would give 40W class A into 10k. Maybe a little more in AB and lower load, but Pete said he thought 5k was already low for SE.
To get more power, you would need more volts but beware the rating for g2. In triode, 1kV would be safe, 1100V is pushing it.
75W triode is very ambitious, maybe in AB2?
If you're very courageous, you can try this one (pentode) Champ 1000 Watt Tube Amp
Thanks
Will go for 1KVA , triode connected
I'd like to keep the driver stages to min. Is cap coupling going to be less efficient than interstage transformer?
Can you get a really good sound without interstage transformer (assuming you are using *really* good coupling caps?
I have a big OPT, and I did a test to find out impedance ratio by applying a precise ac to the primary, working out the turns ratio etc, at 8R sec, the primary impedance worked out to be 1.435K....is that going to be way too low?
Will go for 1KVA , triode connected
I'd like to keep the driver stages to min. Is cap coupling going to be less efficient than interstage transformer?
Can you get a really good sound without interstage transformer (assuming you are using *really* good coupling caps?
I have a big OPT, and I did a test to find out impedance ratio by applying a precise ac to the primary, working out the turns ratio etc, at 8R sec, the primary impedance worked out to be 1.435K....is that going to be way too low?
Is 1.4K one-side or end to end? (We usually specify the OT end to end.) 1.4K end to end is 350 Ohms each tube.
1) Read the 813 data. The suggested load in pentode is 9K to 20K! Triode will be lower, like 2/3rds, or say 6K to 14K. Way above the 1.4K you have.
2) Can you read those curves? or just throwing it down as a challenge? First step: what is the plate resistance as Triode? Looks like 1,600 Ohms. The load on each triode should be 5X to 1X (in push-pull) of the plate resistance. >1600, not 350. Violating this just makes an unhappy amplifier (hot and strained). Plotting load-lines is a lesson which has been taught over and over and over again, do you want a private lesson? I suggest you read Radiotron Designers Handbook. The p-p 2A3 curves are an excellent example, and repeating them on the 813 curves will tell you something.
Taking 1600CT from 1000V goes way over the Pdiss line real quick. This can't be a happy amplifier even for speech/music. Taking 1600CT from 550V gives a perfectly safe amp but swinging only 90V of the 550V you pay for. 6K, 8K, and higher are workable power amps. 1000V and 8K makes about 60 Watts. Somewhere between you can idle at 100mA/tube, work near class A, dissipate 200W for 30% efficiency. Not bad for triode. Going higher voltage and higher impedance makes more power but has to run class AB. (2,000V and 30K makes 150 Watts; but 30K is not a practical wideband audio winding.) (In pentode, twice as much with half the impedance.)
Attachments
Thanks for your load lines. I'm trying hard to understand them. I don't have a copy of RDHv3, but Ill try to find one
Its end to end. How do you get 350R?
And what would be the best B+ voltage to use using the OPT I have (Raa 1.43K)
Can I send you (email) a datasheet from the manufacture of the OTP?
Regards Andrew
Its end to end. How do you get 350R?
And what would be the best B+ voltage to use using the OPT I have (Raa 1.43K)
Can I send you (email) a datasheet from the manufacture of the OTP?
Regards Andrew
Last edited:
What you propose is like a tractor gearbox in a Ferrari. The 813 wants a HIGH "gear ratio". The OT you have is a LOW "gear ratio".
I have reviewed the info from the man who wound the transformer. It was meant for *twelve* KT88/6550, B+ = 550V, rated for 500W for anode load = 820r, with means Va-a = 640Vrms. It has a highly interleaved winding structure, but that does not allow a large change of ratio.
Other info suggests 1k1 to 6r2, which can be 1k43 to 8r at 286 Watts.
There is no way a 800r-1k43 winding can work well at the 10K you propose, which is indeed a decent match for 813.
813's plate resistance is 2K. This OT acts as a near-short to a 2K tube. It can't even see a "sweet spot"; it is far out in the jungle. Distortion will be considerable (not "sweet triode sound"). Damping factor is terrible.
A single pair of bottles, even the large 813, is nothing like the six pairs originally contemplated. There are nearly no single tubes to compete with a 6-pack of KT88/6550.
But the 6C33C may do OK. Plate resistance is nearer 200 Ohms. They are fine looking bottles. Prices are competitive with commodity 813. I pencil 69 Watts out, DF=3+, and nice sane B+ supply (though the driver may need a 500V supply to make the 110V grid swing).
I have reviewed the info from the man who wound the transformer. It was meant for *twelve* KT88/6550, B+ = 550V, rated for 500W for anode load = 820r, with means Va-a = 640Vrms. It has a highly interleaved winding structure, but that does not allow a large change of ratio.
Other info suggests 1k1 to 6r2, which can be 1k43 to 8r at 286 Watts.
There is no way a 800r-1k43 winding can work well at the 10K you propose, which is indeed a decent match for 813.
813's plate resistance is 2K. This OT acts as a near-short to a 2K tube. It can't even see a "sweet spot"; it is far out in the jungle. Distortion will be considerable (not "sweet triode sound"). Damping factor is terrible.
A single pair of bottles, even the large 813, is nothing like the six pairs originally contemplated. There are nearly no single tubes to compete with a 6-pack of KT88/6550.
But the 6C33C may do OK. Plate resistance is nearer 200 Ohms. They are fine looking bottles. Prices are competitive with commodity 813. I pencil 69 Watts out, DF=3+, and nice sane B+ supply (though the driver may need a 500V supply to make the 110V grid swing).
Attachments
Last edited:
Thanks Paul.
Ahh the sweet triode sound...what I love.
Didn't think the OPT would work, but manufacture seemed to indicate otherwise...
Re the 813s , in pentode config, when do they start to draw G1 current ; ie you cant RC the driver (need A2 driver)?
Is it:
a/dependent on the screen voltage, lower screen voltage need G1 current
b/ when the grid goes positive
or both?
Do I really need a OPT Ra-a load of 15k for B+1200V and 750V on G2
?
If operating under these conditions, would RC driver coupling be ok? How about DC cathode driven pentode?
Cheers Andrew
Ahh the sweet triode sound...what I love.
Didn't think the OPT would work, but manufacture seemed to indicate otherwise...
Re the 813s , in pentode config, when do they start to draw G1 current ; ie you cant RC the driver (need A2 driver)?
Is it:
a/dependent on the screen voltage, lower screen voltage need G1 current
b/ when the grid goes positive
or both?
Do I really need a OPT Ra-a load of 15k for B+1200V and 750V on G2
?
If operating under these conditions, would RC driver coupling be ok? How about DC cathode driven pentode?
Cheers Andrew
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.