Hello everyone!
I have been looking at the EL509/519 datasheets, and they state the maximum anode voltage for cold and peak conditions, but I can't find the constant values. That's probably because the tubes were designed to work in pulsed mode, but it would be great to have more information.
Thanks!
Jose
I have been looking at the EL509/519 datasheets, and they state the maximum anode voltage for cold and peak conditions, but I can't find the constant values. That's probably because the tubes were designed to work in pulsed mode, but it would be great to have more information.
Thanks!
Jose
Although in the data for PL519 (EI-Niš, new tubes) they write different, in practice I met that they work in HF-linear (for radio-amateur devices) amplifiers with plate voltages between 1.3kV (full excitation) and 1.5kV (no excitation ). In audio amplifiers (push-pull in class AB1), I tried them at an plate voltage of Ua~800-850V with a stabilized voltage at G2=225V.
Very useful information, thank you! I'm looking to use around 550V between anode and cathode, hopefully it will be OK.
Svetlana says "DC plate voltage 900 V". While they were selling it for audio, they add it "is designed to be a direct replacement for any 6KG6 or EL509."
http://ekladata.com/Pw1_EIWx_5D-1f2huY5k40HrDX8/el519.pdf
At 140Vg2 they will pass >500mA@120Vp. In class A, 500V 70mA will work. In B you can plot almost arbitrary lines. 900V 550mA is 780V swing and 214W out, and you can just barely stay inside a 35W Pdiss, if nothing goes wrong. I have never seen a plan so bold. 100W/pair is in sight if you do all your sums and no major accidents. (900 Volt accidents are no fun...)
http://ekladata.com/Pw1_EIWx_5D-1f2huY5k40HrDX8/el519.pdf
At 140Vg2 they will pass >500mA@120Vp. In class A, 500V 70mA will work. In B you can plot almost arbitrary lines. 900V 550mA is 780V swing and 214W out, and you can just barely stay inside a 35W Pdiss, if nothing goes wrong. I have never seen a plan so bold. 100W/pair is in sight if you do all your sums and no major accidents. (900 Volt accidents are no fun...)
I used PL519s at about 1100 V DC and about 2100 V peak in this design https://home.kpn.nl/verwa255/esl/amp.zip and they didn't explode or flash over. @Lampie519 may have experience with long-term reliability at high voltages, if I remember well he also uses them in his ESL direct-drive amplifiers.
Last edited:
Thank you guys for your replies, I have now enough information to proceed. My goal is to get around 18~20W in an UNSET configuration, and simulations show the best way is with high voltage and ~6K transformers. Nothing higher than 600V anode-to-cathode.
600V is "a walk in the park" for the PL519 tubes....
The tubes in my direct drive amps are more the 10 years in service and are still going strong...
In here they will not get more then 4KV each.
I have not yet tried them with higher voltages but i would not be surprised if they go up to 7KV in a stable amp.
They just look ugly after some years as they get black stains everywhere.
Good luck with your amp (18W is optimistic though)
The tubes in my direct drive amps are more the 10 years in service and are still going strong...
In here they will not get more then 4KV each.
I have not yet tried them with higher voltages but i would not be surprised if they go up to 7KV in a stable amp.
They just look ugly after some years as they get black stains everywhere.
Good luck with your amp (18W is optimistic though)
Err, 4KV territory is a "no-go" zone for me, I want to live 😳
On a more serious note, yes, 20W from an EL509 on a typical SET is probably out of reach, I mean with decent distortion figures. I'm planning to use Tubelab's UNSET, just to try something different. Simulations show that up to 22W is possible before clipping, I hope to get 18 at least. Need to buy a 30W 5K OPT though, looking at Toroidy at the moment.
Well, until I get the OPT, simulations is all I can do 🙂. I can guarantee you the time for burnt components will come quickly after that!
UnSET gets close to 50% efficiency, so for 22W you'd need to burn up about 45W in the output tube.
I reached 22W unSET feeding 395V @ 140mA through a 1k5 load. I don't remember how much V was on the cathode but the tube was probably dissipating 45W.
Serious red plate territory for a E130L but a '519 wouldn't glow.
How long it would last is a different matter.
What does the simulation say about these operating conditions?
I reached 22W unSET feeding 395V @ 140mA through a 1k5 load. I don't remember how much V was on the cathode but the tube was probably dissipating 45W.
Serious red plate territory for a E130L but a '519 wouldn't glow.
How long it would last is a different matter.
What does the simulation say about these operating conditions?
The folks that built the UNSET with sweep tubes 26HU5 and 26LW6 (similar to EL509) seemed to settle on reflected loads of 1.5k to 3k, with B+ voltage below 500V. Does your simulation show better results with a 5K load; at what B+? What driver tube are you using in your simulation? I’m well advanced with my UNSET PCB population but I’m temporarily stuck on the operating conditions for the driver tube that I plan to use (12gn7a on input/driver and 12LW6 output tubes with 3k load and B+ around 450 V).
Could you post your raw simulation files for us to play with? (As you know I’m trying to learn LTSpice.😁)
Attached the simulation file. The comments in blue are a bit out of date, but they show the trend anyway. I tried 3K3, 5K and 6.5K OPTs, and lower impedance/high currents result in a much bigger THD, which matches the usual practice. I do not like big currents, I prefer high voltage low current.
I have tried 3K5 with 430V B+, around 340V anode to cathode, 122mA Iq, I get 18.33W simulated, 5% THD. I get much better results using higher voltage and primary impedances, 5K (540V B+) 6.5K(640V B+).
I got feedback from Toroidy, unfortunately ~35 days to deliver the transformers, they are very busy.
When I find the time (🥴) I'll throw together a higher voltage supply to test the big 35W+ tubes.
I have a pair of heavy 5k9 OPT's waiting for some action.
I have a pair of heavy 5k9 OPT's waiting for some action.
Simulations showed even better results using 6.5K and 640V B+.
Anything we could do to help you find time? 😁