I use Chinese tubes in the design phase of most of my amps. I do this because I tend to test my designs far in excess of the normal operating conditions. I have Chinese 845's (Shuguang) in my amp, and I have run them as hard as 1050 volts at 120 mA. I use my old RCA's sometimes for critical listening, but since these are worth crazy money, I have never cranked them past 70 mA.
After having said all of that, I have never tried Chinese 811's or 813's. I have some old military surplus versions of each.
I personally think that the 6HV5 can be made to work. but I haven't had the time to make one glow yet, much less make any music.
After having said all of that, I have never tried Chinese 811's or 813's. I have some old military surplus versions of each.
I personally think that the 6HV5 can be made to work. but I haven't had the time to make one glow yet, much less make any music.
I think the 6HV5A will work in the sense of delivering volts and amps, I just don't know if it'll sound nice enough to make it worth the effort. I've also heard that they are prone to shorting. I suspect this may be due to the close grid to cathode spacing making it sensitive to any cathode flaking. The GE tubes are supposed to have a diffusion bonded cathode coating that makes them more reliable, but whether this is actually the case or just marketing bally-hoo remains to be seen. Operating them at lower DC voltages rather than pulsed voltages of several kV with high current will probably also make a big difference, as the cathodes should take less of a beating. Anyway, I'd be happy not having to deal with a directly heated tube like the 811A just yet (still more iron), so I hope the 6HV5As work for me.
The Chinese 813 in triode didn't seem to like plate voltages much over 980 Vdc (90 ma) in my tests. Driving an 1628se, eight ohm load on the four ohm tap, anything higher resulted in a long series of low level upper harmonics. At around 950 Vdc they just turn off and leave a very quickly decaying monotonic signature. I've bee too busy with work to persue it but it looks like a builder.
At the time I was testing the old 1628se's suitability as a 10K:8 OPT and didn't experiment much with operating points. Minimizing higher harmonics at the expense of increased second was about it. Without notes I don't recall it objecting to lower voltages. I think UL worked very well too but since triode provides lower Zout with more power than I'll need didn't look into it deeply. And no, I haven't tried a US yet. Too much frenzy on E-pay. This pair of Chinese 'Electron Tube' were as I recall about $60 bucks matched from around your neck of the woods.
THD spectrum at 1 and 4 watts (B+ 990 Vdc, Vg -98 Vdc, CCS-loaded & LED biased 6c45-pe driver) below. No harmonic higher than the third above -95 dB isn't too harsh a price for ~0.9% second.
THD spectrum at 1 and 4 watts (B+ 990 Vdc, Vg -98 Vdc, CCS-loaded & LED biased 6c45-pe driver) below. No harmonic higher than the third above -95 dB isn't too harsh a price for ~0.9% second.
Attachments
I got 4 "untested" 813's on Ebay for $30. One did not need to be tested, it was obviously toast. The other 3 work. I have since found 2 more in my warehouse. I have them on my list of tubes to experiment further on.
I noticed the 60 Hz and all of its relatives on your plots. Also visible is the IM products of the 60 and 120 Hz on either side of the 1KHz and 2KHz tones. I chased a similar situation when I was developing my TubelabSE amplifier. This is caused by AC filaments on the DHT's. I could near it as a vague fogginess on single vocalists, and my hearing is not that good. I found the only cure is DC filaments. This is in direct conflict to the accepted norm that DHT's sound better on AC. YMMV.
My brief experiments with the 811A showed this was a bigger problem. Probably due to the tubes high gain.
I noticed the 60 Hz and all of its relatives on your plots. Also visible is the IM products of the 60 and 120 Hz on either side of the 1KHz and 2KHz tones. I chased a similar situation when I was developing my TubelabSE amplifier. This is caused by AC filaments on the DHT's. I could near it as a vague fogginess on single vocalists, and my hearing is not that good. I found the only cure is DC filaments. This is in direct conflict to the accepted norm that DHT's sound better on AC. YMMV.
My brief experiments with the 811A showed this was a bigger problem. Probably due to the tubes high gain.
I made a pair of 811A monoblocks a while ago.
The main issue was to feed them with 6,3V 4A clean DC, I did it the hard way with iron core inductors and over 100 000uF in a LCLC-configuration.
With about 400V plate-cathode, +24V on the grid and 5k plate load I got 12W or so on the output.
12 quite good sounding watts too!
Due to it´s high plate resistance 811A works more like a pentode than a regular triode, so NFB was necessary to tame the output impedance.
Unfortunately my amps had some PSU problems that I couldn´t solve (back then, now I know better...) so I tore them down and reused the parts in other projects.
Considering the size, weight and look of those amps I don´t miss them anyway...
BTW I used direct coupled EL86 cathode followers to feed the 811´s. If I´d do it again I´d use mosfets instead.
The main issue was to feed them with 6,3V 4A clean DC, I did it the hard way with iron core inductors and over 100 000uF in a LCLC-configuration.
With about 400V plate-cathode, +24V on the grid and 5k plate load I got 12W or so on the output.
12 quite good sounding watts too!
Due to it´s high plate resistance 811A works more like a pentode than a regular triode, so NFB was necessary to tame the output impedance.
Unfortunately my amps had some PSU problems that I couldn´t solve (back then, now I know better...) so I tore them down and reused the parts in other projects.
Considering the size, weight and look of those amps I don´t miss them anyway...
BTW I used direct coupled EL86 cathode followers to feed the 811´s. If I´d do it again I´d use mosfets instead.
Thanks tubelab. It was an early 813-1628se proof-of-concept run, I hadn't even bothered with hum null pots yet. I'll keep an ear out for the effect you mention though, it's always helpful to have a descrition to correlates with a measurement.
Hi Tweeker, I haven't been able to find GU-13s. Heard about them a long time back but they seem to be unobtanium. Do you know a source?
Hi Tweeker, I haven't been able to find GU-13s. Heard about them a long time back but they seem to be unobtanium. Do you know a source?
Attached are a couple of preliminary circuits for the 6HV5A amp for comment. One drives the ourputs directly with a triode connected 6005 fairly heavily biased (~20mA) while the other circuit uses a 6CM7 dual dissimilar triode for preamp and cathode follower drive. Component values are preliminary pending measurements of the 6HV5As at 650V B+, but I think they''re pretty close to what I'll need. Drive currents should be light for the 6HV5A even with positive grid voltage due to the low drive voltage required.
6HS5 / 6HV5(A) - curves @ positive grid voltage
Hello!
Has anyone characterized tubes like 6HS5 / 6HZ5 with grid driven positive? I've seen the curves of 6HZ5A, but they are only up to 0V
I haven't received my 6HS5's yet, but I'd like to do some paperwork before I start warming-up the iron
Many thanks for any help!
Hello!
Has anyone characterized tubes like 6HS5 / 6HZ5 with grid driven positive? I've seen the curves of 6HZ5A, but they are only up to 0V
I haven't received my 6HS5's yet, but I'd like to do some paperwork before I start warming-up the iron
Many thanks for any help!
yesterday evening I have started some measurement about 6HV5A used like 845 driver.
The schematic was very simple: inductance on anode, Rk = 100ohm, Ck = 470uF, Rg = 75Kohm and output capacitor 2.2uF on 33 or 100Kohm load.
With medium voltage operation (about 600V) these tubes give low sperformances because distortion is high 1.1% at 25Vrms.
The good result are obtain using 980V and 100ohm on catode to keep the power nead the 30w (35w is the maximun allowed), bias about 30mA.
With these value I have got the good 1% at 120Vrms on 33Kohm load.
The freq. response is limited to 50Hz(-3db) with 20H but it will be 10Hz(-3db) using 100H 50mA Sowter 8982.
The voltage amplification is 120/0.7=155x.
The schematic was very simple: inductance on anode, Rk = 100ohm, Ck = 470uF, Rg = 75Kohm and output capacitor 2.2uF on 33 or 100Kohm load.
With medium voltage operation (about 600V) these tubes give low sperformances because distortion is high 1.1% at 25Vrms.
The good result are obtain using 980V and 100ohm on catode to keep the power nead the 30w (35w is the maximun allowed), bias about 30mA.
With these value I have got the good 1% at 120Vrms on 33Kohm load.
The freq. response is limited to 50Hz(-3db) with 20H but it will be 10Hz(-3db) using 100H 50mA Sowter 8982.
The voltage amplification is 120/0.7=155x.
Attachments
Interesting thread, These sound a lot like the 6BK4B and 6BK4C that I have (one of each). I have been wondering if they had any practical use aside from generating X-rays. Interestingly the pinout diagram does not show the beam grid so I assume it must be internally grounded.
It sounds like these may require more heroics that I am inclined to pursue but they are interesting bulbs.
It sounds like these may require more heroics that I am inclined to pursue but they are interesting bulbs.
wrenchone said:
2nd schematic of the PDF:
Get rid MJE350 and bias network.
Q2, C1, R4, R14... Pointless at 250V...
The current source is under no stress.
Disconnect "plate" end of C4.
Re-connect C4 to top end of R2.
"IXCP as Mu Follower" version.
Better yet, cut R2 in twain. Tap
your drive signal from the middle.
"IXCP as Anti-Triode" version.
The 6BK4 is a triode (3 elements) designed to operate as a shunt regulator on the secondary side of the flyback transformer at 25KV. I found them virtually useless at plate voltages below 1KV. I included them in a batch of tubes that I sold off a while ago.
The 6HV5 and its relatives are beam triodes (4 elements) designed to operate on the primary side of the flyback transformer at 1 to 4 KV. These can be made to operate below 1KV. I had one hooked up and playing music a while back but it did not sound great. I wired it up directly from the output of a CD player and got enough gain to make sound (maybe a watt or two). I came to the conclusion that this tube has a pentode like output resistance that would benefit from some NFB, but there was not enough gain, so I put everything on hold. I saved a bunch of these guys because I am convinced that they will rock when I find the right circuit.
The 6HV5 and its relatives are beam triodes (4 elements) designed to operate on the primary side of the flyback transformer at 1 to 4 KV. These can be made to operate below 1KV. I had one hooked up and playing music a while back but it did not sound great. I wired it up directly from the output of a CD player and got enough gain to make sound (maybe a watt or two). I came to the conclusion that this tube has a pentode like output resistance that would benefit from some NFB, but there was not enough gain, so I put everything on hold. I saved a bunch of these guys because I am convinced that they will rock when I find the right circuit.
Kempeter, The current souce as presented needs the cascode, as the top element is a lowly 40VJFET. Two years down the road, I would just use a depletion mode mosfet for the whole mess and be done with it. These days I might consider using something like a 6AW8 with the pentode at the input and the triode as a follower.
One of these days I need to breadboard something up with a single 6HV5A at the output, as I have a pair of 10W 8k output transformers looking for gainful employment and some electrophoresis supplies I can use for the various B+ flavors. I have five or six amps to finish first, though, and will pick the low-hanging fruit first, if you know what I mean.
One of these days I need to breadboard something up with a single 6HV5A at the output, as I have a pair of 10W 8k output transformers looking for gainful employment and some electrophoresis supplies I can use for the various B+ flavors. I have five or six amps to finish first, though, and will pick the low-hanging fruit first, if you know what I mean.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.