Greetings all, my name is Steve. I've been a member here for a little while and primarily read things here, so this is a rarity for me.
I've built a few things in the past using HV (PP 807's was my most recent project) so I understand the dangers and costs involved with building big amps.
I want to build a SET Amp using the 808. Right now, here's how far I am:
I have a big old Kenyon transformer that can do EITHER 2000VCT, 2500VCT, or 3000VCT at about 250ma. Based on the datasheet I have for the 808, it looks like at max plate voltage (1500) I'll be using the 2500 taps.
I have two globe 866's that test new and are pretty sweet, as well as a filament transformer with CT and a 2kv+ rating. I know people advise against MV rectifiers, but there's no way I'm not using these. They'll be beautiful.
Have a choke, 12H at 250ma. Here's where it gets fuzzy... I don't know the HV test rating of it. I cannot find data anywhere. All I know is it's pretty big. That being said, what are the disadvantages of putting the choke on the CT of the high voltage? Will this mess with my MV tubes?
The next area of concern is using the 808 class A. I know it's a radio transmitting tube and designed for B or C but I know shishido did it class A if I remember correctly. If it is indeed possible, what would be a good tube to drive it with? a pair of 300b was my first thought? Prefer some kind of tube that will look as pretty as the 866's and the 808.
Curious to hear thoughts.
Steve
I've built a few things in the past using HV (PP 807's was my most recent project) so I understand the dangers and costs involved with building big amps.
I want to build a SET Amp using the 808. Right now, here's how far I am:
I have a big old Kenyon transformer that can do EITHER 2000VCT, 2500VCT, or 3000VCT at about 250ma. Based on the datasheet I have for the 808, it looks like at max plate voltage (1500) I'll be using the 2500 taps.
I have two globe 866's that test new and are pretty sweet, as well as a filament transformer with CT and a 2kv+ rating. I know people advise against MV rectifiers, but there's no way I'm not using these. They'll be beautiful.
Have a choke, 12H at 250ma. Here's where it gets fuzzy... I don't know the HV test rating of it. I cannot find data anywhere. All I know is it's pretty big. That being said, what are the disadvantages of putting the choke on the CT of the high voltage? Will this mess with my MV tubes?
The next area of concern is using the 808 class A. I know it's a radio transmitting tube and designed for B or C but I know shishido did it class A if I remember correctly. If it is indeed possible, what would be a good tube to drive it with? a pair of 300b was my first thought? Prefer some kind of tube that will look as pretty as the 866's and the 808.
Curious to hear thoughts.
Steve
I love to see people starting projects like this.
I do think that 808 is a bit of a problem for audio with its very high anode resistance and mu.
The only way I should think would be in class A2; there is a Japanese bloke who has obtained good results, if I can find a link I'll post it here,
To drive the large amount of grid current, you would need something more suitable than 300B I think. I suppose a transformer would be the answer but I prefer glass, so I would make a super-beefy cathode follower using a power valve with plenty of poke. If it had to be a bit different, a cathode follower built using a QQVO6/40 (sorry forgotten the American number - it's 5xxx} Running the two sections in parallel as beam tetrode cathode follower would perhaps provide the necessary grunt! There are still beefier versions of these double beam tetrodes - YL1071 also has a US number - 8xxx - sorry!
Lastly I did always quite fancy the 806 which looks like the 808 but has a base like an 845; it looks as though it might be a bit less demanding.
Edit regarding the choke, perhaps you know someone who has an insulation tester? Of course it would have to be a high-voltage one and if we are talking of 2kV then we're looking at proper porcelain insulators etc., which might be outside the realm of your choke!
I do think that 808 is a bit of a problem for audio with its very high anode resistance and mu.
The only way I should think would be in class A2; there is a Japanese bloke who has obtained good results, if I can find a link I'll post it here,
To drive the large amount of grid current, you would need something more suitable than 300B I think. I suppose a transformer would be the answer but I prefer glass, so I would make a super-beefy cathode follower using a power valve with plenty of poke. If it had to be a bit different, a cathode follower built using a QQVO6/40 (sorry forgotten the American number - it's 5xxx} Running the two sections in parallel as beam tetrode cathode follower would perhaps provide the necessary grunt! There are still beefier versions of these double beam tetrodes - YL1071 also has a US number - 8xxx - sorry!
Lastly I did always quite fancy the 806 which looks like the 808 but has a base like an 845; it looks as though it might be a bit less demanding.
Edit regarding the choke, perhaps you know someone who has an insulation tester? Of course it would have to be a high-voltage one and if we are talking of 2kV then we're looking at proper porcelain insulators etc., which might be outside the realm of your choke!
Last edited:
The only real disadvantage of using the choke in the return side is that stray capacitances to ground tend to bypass the choke rather than adding to the "real" caps. No big deal usually.
Also, because the MV rectifiers will want to see a choke input, the first choke will see a *lot* of AC voltage across itself. Always a good idea to minimize high voltage peaks as possible. Some folks will add a snubber across the input choke - can't hurt.
All good fortune,
Chris
Also, because the MV rectifiers will want to see a choke input, the first choke will see a *lot* of AC voltage across itself. Always a good idea to minimize high voltage peaks as possible. Some folks will add a snubber across the input choke - can't hurt.
All good fortune,
Chris
Shishido used 350 volts for 15 watts output with 120Ma current.
Only my opinion but I don't think this tube would sound very good at 1500 volts.
You would have to run it at 5-10ma or burn it out. the 808 only has 50 watts plate dissipation!!!
These tubes work best at lower voltages and high current.
the 805 sounds terrible at 1200 volts Single end, 20 Ma, but very good as Shishido uses it at 800 volts and 120-140Ma
I can scan Shishido's circuit if you wish.
Phil
Only my opinion but I don't think this tube would sound very good at 1500 volts.
You would have to run it at 5-10ma or burn it out. the 808 only has 50 watts plate dissipation!!!
These tubes work best at lower voltages and high current.
the 805 sounds terrible at 1200 volts Single end, 20 Ma, but very good as Shishido uses it at 800 volts and 120-140Ma
I can scan Shishido's circuit if you wish.
Phil
Since you haven't said that you actually have some 808s in your possession, please be very cautious if you need to buy them. I can tell you from experience that these are old enough that 50% of what's out there will be either very gassy or have gone to air. Most sellers don't have a clue what the real condition is for any they may have for sale. Buy only with a full guarantee and return privilege. Regular normal tube testers can't test them. Also, the filaments are very fragile and demand very careful packaging for shipment.
Thanks for the replies guys. I have two 808's at this point, both were properly tested before I purchased them and I have checked that the filaments are still good in each, and have lit the filaments as well recently.
Low voltage may be a better option then, I didn't really think of that.
Interesting point on the choke too, I will probably look into that more and add a snubber.
I can't wait to get started on this thing... I also have two 811's laying around, but I don't know if those would be any easier and I'd rather stick with the 808.
Also will look into more options for driving it. I too am a fan of glass, and would prefer to not have an interstage transformer in it at all.
Steve
Low voltage may be a better option then, I didn't really think of that.
Interesting point on the choke too, I will probably look into that more and add a snubber.
I can't wait to get started on this thing... I also have two 811's laying around, but I don't know if those would be any easier and I'd rather stick with the 808.
Also will look into more options for driving it. I too am a fan of glass, and would prefer to not have an interstage transformer in it at all.
Steve
Nice project!
I´ve been collecting parts for an 808 SE amp for several years, changing my mind countless times along the way and so far coming up with nothing but full storage shelves and empty pockets...
I picked 2C22 (~1/2 6SN7 with dual top caps) as input tubes and RK60 (5R4 with dual top caps) as rectifiers, mostly because of the visual impact they make together with the 808.
As HolloState says, 808 is not the easiest tube to find. 811A might work as a substitute with some modifications (6,3V vs 7,5 on the filaments and 24V vs 36 on the grids).
Those kilovolts you see in the datasheet refers to class B operation, not very useful in SE class A2. 350-450V, 100-120mA and 2,5-3,5k is more like it, resulting in 10-15W output power.
I´ve been collecting parts for an 808 SE amp for several years, changing my mind countless times along the way and so far coming up with nothing but full storage shelves and empty pockets...
I picked 2C22 (~1/2 6SN7 with dual top caps) as input tubes and RK60 (5R4 with dual top caps) as rectifiers, mostly because of the visual impact they make together with the 808.
As HolloState says, 808 is not the easiest tube to find. 811A might work as a substitute with some modifications (6,3V vs 7,5 on the filaments and 24V vs 36 on the grids).
Those kilovolts you see in the datasheet refers to class B operation, not very useful in SE class A2. 350-450V, 100-120mA and 2,5-3,5k is more like it, resulting in 10-15W output power.
You'll probably want a nice clean filament supply for the 808. The mu is pretty high and a little bit of ripple will really be audible. 30 watts for a filament is more typical of tubes with 100 watt Pd like the 845 or 211.
I did a bit of playing with a regulator chip from Micrel, the MIC5156 MIC5156/5157/5158 Super LDO™ Regulator Controller. It drives any large mosfet as a pass device for the regulation, the only drawback is it needs a second higher voltage to properly drive the mosfet gate.
Somewhere I used to have a page bookmarked that was in japanese but had tons of schematics for oddball tubes. They had a few designs based on the 808 and had a switch for running with an 812 as well. I was going to build one but the numbers they posted for power output vs distortion weren't too impressive.
I did a bit of playing with a regulator chip from Micrel, the MIC5156 MIC5156/5157/5158 Super LDO™ Regulator Controller. It drives any large mosfet as a pass device for the regulation, the only drawback is it needs a second higher voltage to properly drive the mosfet gate.
Somewhere I used to have a page bookmarked that was in japanese but had tons of schematics for oddball tubes. They had a few designs based on the 808 and had a switch for running with an 812 as well. I was going to build one but the numbers they posted for power output vs distortion weren't too impressive.
It's true, the ripple will make itself felt. With a carefully designed Regulator, the feedback for the filament current does not get mixed-up with the cathode current. I believe this is best implemented with discrete transistor design - which can eliminate the need for the noisy bandgap reference-voltages found in IC regulators.
I have a purpose-built Regulator for the RCA-808. It's a discrete design with ultra-low self-noise, and has a control-loop which makes no contact with the filament. Available as self-assembly kit, DIY friendly. Plenty of DIYers on our forum have expressed their pleasure at the difference to the sound.......
.
Attachments
Further to the link I posted at No. 14, I refrained from commenting on the designer's input valve choices on the grounds that he is obviously jolly clever and no doubt has reasons for his choices.
I would not use 12AT7 and especially not 12AU7 as he has in his designs. SImple alternatives would be 12AY7/6072 and 6SN7. I see however that the 838 design which is perhaps relevant to this thread has 77v rms (+40dB) feeding the grid of the 838. I once built a very low distortion driver stage providing this output - I measured 0.063% this driving a 30k pad. The stage used a 7N7 (what else!) followed by a pair of E810F (7788) pentode cathode followers. The stage did require +ve and -ve HT supplies but was designed for DC coupling to the output valves (in this case triode strapped YL1071).
Sadly this project never got past the prototype stage. Such is life
I would not use 12AT7 and especially not 12AU7 as he has in his designs. SImple alternatives would be 12AY7/6072 and 6SN7. I see however that the 838 design which is perhaps relevant to this thread has 77v rms (+40dB) feeding the grid of the 838. I once built a very low distortion driver stage providing this output - I measured 0.063% this driving a 30k pad. The stage used a 7N7 (what else!) followed by a pair of E810F (7788) pentode cathode followers. The stage did require +ve and -ve HT supplies but was designed for DC coupling to the output valves (in this case triode strapped YL1071).
Sadly this project never got past the prototype stage. Such is life
Here's how it turned out.
Sorry this took me three and a half years to build. Anyway, here it is. Huge thanks to this forum and to Rod for his DHT regulators. The thing sounds great.
You guys were all really helpful and I can't wait to build whatever comes next.
PS, yeah, don't touch the caps. Seriously.
Sorry this took me three and a half years to build. Anyway, here it is. Huge thanks to this forum and to Rod for his DHT regulators. The thing sounds great.
You guys were all really helpful and I can't wait to build whatever comes next.
PS, yeah, don't touch the caps. Seriously.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.