The differences between mechanical resonances due to the internal structures. It's funny how tubes can mechanically "sing" at certain frequencies. Try loading a power tube with a "non singing load" (not a transformer), such as a power resistor or a light bulb and do a frequency sweep.
My experience with GM-70 is quite good. I'm able to get 32W out of it (1050V 95mA), bias is fixed but variable (it "slides" with B+ so there is no concern about disipation).
Input tube is 3A5, driver 4P1L with IT to GM-70. I'm very very satisfied with the sound. I'll stick with GM-70 for a long time.
Next "modification" will be even higher B+. I'm aiming at 1200V 80mA. I have all the parts but don't have time to try it. This year for sure.
GM-70 FTW!
Input tube is 3A5, driver 4P1L with IT to GM-70. I'm very very satisfied with the sound. I'll stick with GM-70 for a long time.
Next "modification" will be even higher B+. I'm aiming at 1200V 80mA. I have all the parts but don't have time to try it. This year for sure.
GM-70 FTW!
As far as new 845 production, I can say I have a pair of 845s used as a daily driver and have not had a failure with about two years of use.
My guess is about four hours of daily use on average.
Amazon.com: 2pcs Psvane 845 HiFi-series Vacuum Tubes Matched Pair: Home Audio & Theater
I am running as a low power (about 10 watts). Mostly a comment about filament and getter performance.
My guess is about four hours of daily use on average.
Amazon.com: 2pcs Psvane 845 HiFi-series Vacuum Tubes Matched Pair: Home Audio & Theater
I am running as a low power (about 10 watts). Mostly a comment about filament and getter performance.
GM-70 were made for military. In USSR there was no market, it was planned economy, so specs were not to impress potential buyers, but to avoid jail for failure, so you can assume that GM-70 would take some abuse, and live much longer in audio applications that you can read in datasheets.
It's funny how tubes can mechanically "sing" at certain frequencies.
This can be solved by Herbie Tube Dampers use the model UltraSonic RX-65 the same used in 6C33
years ago, i ran a sweep tru ecc83, some tones could be heard inside. (but with big amplitude though)
This is due to a phenomenon called magnetostriction and well defined: Magnetostriction - Wikipedia
OTOH, the claim of audio coming directly out of the tubes' envelopes, and even more the effect of it's damping by obscure rings (made from polymerized snake oil?) lacks substantial evidence.
Best regards!
OTOH, the claim of audio coming directly out of the tubes' envelopes, and even more the effect of it's damping by obscure rings (made from polymerized snake oil?) lacks substantial evidence.
Best regards!
Lately I did a sweep on a triode strapped 6P45S that rang so well. In fact, it did almost like an OPT coil. Just another "accidental" discovery.
They were used in modulators of AM transmitters and large stadium amplifiers. Quality and QC is pretty variable, but they are cheap, and when new they all sound pretty good.
They actually do not live much beyond their rated lives - I've been using them for about 7 years now and have been through about 10 pairs in that time. I keep loose track of run times and they seem to be worn out around 1200 hours, I did have one pair of coppers that went nearly 1500 hours before I felt compelled to replace them - how do you know? They just start to sound bad..
Emission drops rapidly beyond 1000 hours in most of them.
Yes, GM-70 except modulators of military AM transmitters were also used in Tu-600 amps that always were Government property, and had to be replaced every 1,000 hours. And as all Soviet tubes except few designed by Telefunken engineers, had spread of parameters wider than barn gates. QC was really strict, unlike for tubes that were for sale for usage by civilians. They were often garbage.
Hi,
Site QRO-Parts.com GS35b GU84b GU74B GI7b 4CX800a had GMI-11 they say this tube last 25 years in the Sukhoi-29 radar, this is correct?
How many is the GU-72 time life in audio Class A1 10W?
I doubt that somebody would risk so much to leave the tube for 25 years in a radar instead of replacing each 300 hours of usage, according to Reglament.
Only if during 25 years if worked 300 hours total.
I have no experience with Gu-72. I have experience with Gu-50 though, that outlive guaranteed time many times. When "used" measure like new.
Do you preheat? What's your operating point? Anode disipation?
My pair is like new and it's been 4 years. Ok, I don't listen that much because I have a little spare time, but I preheat them from 15 to 30minutes before I turn on B+. Before I turn them off, I turn of B+, wait for about 10minutes and then turn off the amp.
No, I don't preheat and B+ is applied with the filament and bias voltage. The filaments are heated by CCS which takes a few seconds to come all the way up.
The filaments are straight tungsten according to the people I have talked to in Russia about them, so stripping should not be an issue.
1kV, 90 - 120mA depending on whether copper plate or graphite, 7K plate load.. Class A1 operation only, the driver is too wimpy to provide any significant grid current. (90 - 100W copper, 110W graphite typically)
Fixed bias is employed and the driver is a triode connected D3A into a 5K 1:1 amorphous core interstage transformer.
I claim 20W but that is highly dependent on tubes and operating point used.
I average something like 30 hours a week of listening these days. I've never had a filament or catastrophic failure. I usually can tell when they need replacement without any special effort. Every tube is different, but on some they no longer maintain the set operating current and need more and more frequent adjustment until you can no longer adjust them.. Others stolidly plod along at close to set current for their entire lives. Usually the deterioration in sound quality starts when the getters are about 50 - 60% of their initial size.
The filaments are straight tungsten according to the people I have talked to in Russia about them, so stripping should not be an issue.
1kV, 90 - 120mA depending on whether copper plate or graphite, 7K plate load.. Class A1 operation only, the driver is too wimpy to provide any significant grid current. (90 - 100W copper, 110W graphite typically)
Fixed bias is employed and the driver is a triode connected D3A into a 5K 1:1 amorphous core interstage transformer.
I claim 20W but that is highly dependent on tubes and operating point used.
I average something like 30 hours a week of listening these days. I've never had a filament or catastrophic failure. I usually can tell when they need replacement without any special effort. Every tube is different, but on some they no longer maintain the set operating current and need more and more frequent adjustment until you can no longer adjust them.. Others stolidly plod along at close to set current for their entire lives. Usually the deterioration in sound quality starts when the getters are about 50 - 60% of their initial size.
Very interesting procedures, I suppose it will help any tube.
Thanks for inform :-DD
I suspect that many GM-70 you buy already served their 1,000 hours in military transmitters. When I bought GK-71 tubes, they were some loose, some in original boxes. All looked like new. But while experimenting, I had to set different bias voltages for "loose" tubes. After finished the amp, I installed 2 from boxes. Bias voltage got a bit higher, and EQUAL for both tubes! I did not select them. It is rare for Soviet made tubes except 4P1L and alike, but it is the fact as is, tubes from boxes were as if matched, and had higher emission.
So, my conclusion was, when they age, they lose emission differently, that's why used tubes have wider spread of parameters.
I may be wrong, though.
I drive GK-71 in A2, with 800V B+, loaded on 5K, getting 45W per tube.
So, my conclusion was, when they age, they lose emission differently, that's why used tubes have wider spread of parameters.
I may be wrong, though.
I drive GK-71 in A2, with 800V B+, loaded on 5K, getting 45W per tube.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.