Therefore Fermi was a really cool dude......! My favorite line he said was back when he was about to place the first atomic pile into a sustaining chain reaction and someone said.... What if something goes wrong? Fermi's reply was "I will put on my overcoat and simply walk away". In reality I doubt that he would have even gotten to his overcoat if something had gone wrong that day. Those guys that did that experiment really had guts... some of them really thought it was going to go boom, and I suppose it could have were it not for Fermi's insight and brilliance into how it shold behave.
Mark
I am interested in PL/EL 509 push pull schematics if anyone has some they can send?
Regards
Pete
Check out Audio Reality book from Transcendent sound. It has Super Compact project that give you >100W from a pair of EL509.
http://http://www.transcendentsound.com/Transcendent/Books.html
http://http://www.transcendentsound.com/Transcendent/Books.html
Back in the 70th I worked at Philips and was known to be a tube freak. So, when nobody else even looked at tubes to be worth anything I was very exited when my boss handed me a box with 100pcs EL509 and told me I can have them because otherwise they just would get trashed with the rest of the tube stock cleared out. Shortly after, I left Philips to start my own company. In the beginning I had not many customers. I had the time to investigate and build all kind of stuff and could afford to play with the the EL509s. I thought, the next best thing to a beefy triode amp must be a beefy quasi-triode. And 100 x 30W pa was nothing to sneeze at.
First thing, I had to find pairs. A crude attempt did not bring much, measurement results where all over the place, huge tolerances. A bit disappointed I went on to determine the max Ua+Ug2 at various settings. The first tube I tested did not mind 300...400 450....WOW, awsome 500V!!!. All exited I stuffed in the next tube, then the next...until.... first a tiny sizzle...than ...firework...just a bit above 400V. Ok, bad tube, try some more....next tube... and next.... it seemed to go on for ever......all passed up to (otherwise limited)500V Ua+Ug2 at 50mA Ia+Ig2. After a couple of days swapping...adjusting...watching...and nothing happening...I got tired and decided that at least 400V should be a safe voltage on a+g2. Next, may be "burn in" was needed to stabilize parameters, but how to burn in 99 tubes? I ended up mounting 20 sockets in a row , heaters in seria a.s.o. Now I wold only have to wait a week before I could take serious measuments. Plenty of time to cobble up some "precision" gear for the task. Well, it was not exactly precision, but it was at close as I could get under my limited resources...ferro-resonant stabilizer...variac....transformer/ graetz/crcrcr...more of the same with a negativ attitude coupled to some precision-pots...and off course all the meters I could lay hands on. With all those efforts made I was pretty sure that measurement results are reliable and I will end up the proud owner of 49 pairs of fine tubes lasting at least the 6 lifes I still had to look forward too. So I went measuring...adjusting...taking readings...-1V more on the grid....adjusting..taking readings...endlessly...hours...and hours...Finally I figured I gotta take bigger steps in life, 1 volt wont cut it. Ok, things went faster
after quintipling but still not exactly fast. I think I was allready into the third week and 80+ tubes still waiting to get measured. Time to take a break and look at the results so fare.Common wisdom bigger is better is certainly true when it comes to drawing Ia/-Ug curves on mm-paper. I propable drawed the most readable and real curves ever produced ond paper that covered half of my living rooms floor. And I drawed them on top of each other...and no...they where not beautiful to look at...and no...non of them matched...not even remote...The more dots I connected and the farther the drawings went the more desperate I became...this tube is not what I hoped fore...Eventually I just had to admit, all the effort was a waste of time...wasted on the wrong type of tube....I was frustrated and angry and in revange I slaugthered some, ripped the guts out...It was a eyeopener, certainly all became clear...uncentered...crudely built out of chickenwire. Well, what would have been reasonable to expect from a tube built to be a switcher?
This episode was one of the crucial steps in my learning curve. I trashed the whole bunch of them and never regretted it.
First thing, I had to find pairs. A crude attempt did not bring much, measurement results where all over the place, huge tolerances. A bit disappointed I went on to determine the max Ua+Ug2 at various settings. The first tube I tested did not mind 300...400 450....WOW, awsome 500V!!!. All exited I stuffed in the next tube, then the next...until.... first a tiny sizzle...than ...firework...just a bit above 400V. Ok, bad tube, try some more....next tube... and next.... it seemed to go on for ever......all passed up to (otherwise limited)500V Ua+Ug2 at 50mA Ia+Ig2. After a couple of days swapping...adjusting...watching...and nothing happening...I got tired and decided that at least 400V should be a safe voltage on a+g2. Next, may be "burn in" was needed to stabilize parameters, but how to burn in 99 tubes? I ended up mounting 20 sockets in a row , heaters in seria a.s.o. Now I wold only have to wait a week before I could take serious measuments. Plenty of time to cobble up some "precision" gear for the task. Well, it was not exactly precision, but it was at close as I could get under my limited resources...ferro-resonant stabilizer...variac....transformer/ graetz/crcrcr...more of the same with a negativ attitude coupled to some precision-pots...and off course all the meters I could lay hands on. With all those efforts made I was pretty sure that measurement results are reliable and I will end up the proud owner of 49 pairs of fine tubes lasting at least the 6 lifes I still had to look forward too. So I went measuring...adjusting...taking readings...-1V more on the grid....adjusting..taking readings...endlessly...hours...and hours...Finally I figured I gotta take bigger steps in life, 1 volt wont cut it. Ok, things went faster
after quintipling but still not exactly fast. I think I was allready into the third week and 80+ tubes still waiting to get measured. Time to take a break and look at the results so fare.Common wisdom bigger is better is certainly true when it comes to drawing Ia/-Ug curves on mm-paper. I propable drawed the most readable and real curves ever produced ond paper that covered half of my living rooms floor. And I drawed them on top of each other...and no...they where not beautiful to look at...and no...non of them matched...not even remote...The more dots I connected and the farther the drawings went the more desperate I became...this tube is not what I hoped fore...Eventually I just had to admit, all the effort was a waste of time...wasted on the wrong type of tube....I was frustrated and angry and in revange I slaugthered some, ripped the guts out...It was a eyeopener, certainly all became clear...uncentered...crudely built out of chickenwire. Well, what would have been reasonable to expect from a tube built to be a switcher?
This episode was one of the crucial steps in my learning curve. I trashed the whole bunch of them and never regretted it.
Last edited:
Of course if you use a design that necessitates perfectly matched tubes at all cost, you will end up throwing everything in the bin.
Personally I design my circuits to make allowances for the fact that bits of metal in a glass tube won't be identical from one piece to the next. That's all part of the engineering process...
I've never traced a curve nor measured a tube in that detail in my life ad have turned out dozens of amplifiers from an eclectic array of tubes over time.
I have lots of new old stock EL509 tubes and mine certainly wont be going in the bin!!!
Personally I design my circuits to make allowances for the fact that bits of metal in a glass tube won't be identical from one piece to the next. That's all part of the engineering process...
I've never traced a curve nor measured a tube in that detail in my life ad have turned out dozens of amplifiers from an eclectic array of tubes over time.
I have lots of new old stock EL509 tubes and mine certainly wont be going in the bin!!!
And how would you "design" to tolerate the untolerable? Waste power in huge cathode resistors, or would you rather use a transformer with a huge airgap?
There are limits in what I find to be tolerable....
I have built some. Below is the first one. It is pentode connected with fixed bias.
All transformers are from Italian Inmadout.com. The OPT is 2k and the output power is some 170...190 W.
I have tested this circuit with PL519 and soviet 6P45S. Both give equal performance.
There is a modification I have done compared to the diagram showed.
The screen grid current is quite high with full power and therefore I added a NFET to buffer the 0D3 stabilizer and reduced the screen grid resistor (680) to 47 ohms.
An externally hosted image should be here but it was not working when we last tested it.
I have also done a 100 W Circlotron and one 100 W PP with JJ EL509, concertina phase splitter and 4k to 8 opt.
I can show the schematics if you are interested about these too.
I think you see problems where none exist, by your reckoning its impossible to build an EL509 based amplifier and clearly that is not the case.
The detail is in the design of the circuit. I see no such problems and certainly none that are insurmountable in the way you suggest.
This is a 100...120 W PP version with JJ EL509 tubes and Polish Indel transformers.
I have not tested this with "real" PL519.
The voltage amplifier and phase splitter has been done with soviet 6F12P "super tube".
I like this tube a lot. The triode section has mu of 100 and gm = 19.5 mA/V.
Pentode side has the same gm.
Below are the amplifier and PS.
My EL509 Circlotron is shown here:
http://www.diyaudio.com/forums/tubes-valves/231437-100-w-class-circlotron.html
I have not tested this with "real" PL519.
The voltage amplifier and phase splitter has been done with soviet 6F12P "super tube".
I like this tube a lot. The triode section has mu of 100 and gm = 19.5 mA/V.
Pentode side has the same gm.
Below are the amplifier and PS.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
My EL509 Circlotron is shown here:
http://www.diyaudio.com/forums/tubes-valves/231437-100-w-class-circlotron.html
Last edited:
If someone wants to use this tube in pentodeamps and especially if all he wants is power, this tube will deliver. But if someone plans to use this tube in the way I talked about, connected in QUASITRIODE he would have to acceptamplificationfactors that can be anything between approx 2,5 and almost 6. Sorry to say, but this is really not what I think of being tolerable.
TV horizontal sweep tubes were used here...... McIntosh MC 3500 MC 3500 Mono Amplifiers Used in Recording Studio | eBay
as pentodes, sure...
If someone wants power instead of the usually quite nice distortion spectrum of a quasitriode it doesn matter how useful they are in triode-connected.
Last edited:
I never said that. I simply said that my IaIg2/-Ug1 measuring results
revealed that the measured EL509 where absolutely disappointing and no way they are worth using in the quasitriode PP amp I had in mind.
Mind you, at the time there where still plenty of much better choices avaiable and even I got them for nothing I was much better off buying some other tubes. As to "the detail is in the design" are you suggesting that a highly unlinear and high tolerance EL509 will giv you results as good as one of more favourable tube-tubes is capable off just by using some myracle "design details"? If so, I would like to hear about what that "details" are (other than plenty of feedback).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.