Hey Everyone,
New the the forums here but definitely not new to building guitar amps. Currently I'm working on converting one of my old single ended builds to use a kt120 instead of the EL34 that was in there before. I've ordered some new transformers and thought I had everything sorted out until I read that due to the higher inter-electrode capacitance present in the kt120 it might be necessary to have a driver stage. This is one area of tube amp design I basically have no experience with, so I wanted to see what you all would recommend. The pre is a pretty standard trainwreck express minus the PI obviously but I don't know if the two 12AX7's would have enough gain to drive this thing to full power. If not, I wanted to see how I might go about designing a proper driver stage and what tube would be most appropriate. B+ will be around 400V and I plan on biasing it up close to the dissipation limit as I'm really curious to see what this thing can do, so I want to make sure I have the drive capabilities necessary to push it into clipping or close to it.
Thanks in advance.
Necessity of Driver stage for KT120
Moderators: pompeiisneaks, Colossal
Re: Necessity of Driver stage for KT120
I don't know if I can help much, but I'd like to see a diagram of the existing circuit so I'll (we'll) know what you're starting with.
D
D
There are no stupid questions, just stupid people.......
Re: Necessity of Driver stage for KT120
Hi,
12AT7 tubes are needed in the PI circuit.
Have a look at this page:
http://www.tube-town.de/ttforum/index.p ... 03.15.html
The thread contains a working KT120 power amplifire schematic.
You must register, to see the schematic.
12AT7 tubes are needed in the PI circuit.
Have a look at this page:
http://www.tube-town.de/ttforum/index.p ... 03.15.html
The thread contains a working KT120 power amplifire schematic.
You must register, to see the schematic.
-
Leo_Gnardo
- Posts: 2584
- Joined: Thu Sep 27, 2012 1:33 pm
- Location: Dogpatch-on-Hudson
Re: Necessity of Driver stage for KT120
Single-ended, also self-biased? The output tube will "find its own" bias point as long as you don't have an absurdly low value resistor on the cathode. If you can select an output tube with a low emissivity, so much easier for it to settle into its bias zone without running away red plating.
You just need one hi-current triode to drive your KT120 - think about the low-gain side of a 12DW7 (it's pretty much a 12AU7) and you can stick with your typical 9-pin drive tubes and socket wiring. The hi gain side of the 12DW7 you can use for your preamp. JJ has had their 12DW7 copy around for a while, I think they call it ECC832.
What transformers have you selected?
You just need one hi-current triode to drive your KT120 - think about the low-gain side of a 12DW7 (it's pretty much a 12AU7) and you can stick with your typical 9-pin drive tubes and socket wiring. The hi gain side of the 12DW7 you can use for your preamp. JJ has had their 12DW7 copy around for a while, I think they call it ECC832.
What transformers have you selected?
down technical blind alleys . . .
-
David Root
- Posts: 3540
- Joined: Fri Aug 04, 2006 3:00 pm
- Location: Chilliwack BC
Re: Necessity of Driver stage for KT120
12BH7 might be worth a look, in self biased cascode mode. Just speculative on my part, never tried it, but it looks intriguing in TubeCad.
You do not have the required permissions to view the files attached to this post.
-
potatofarmer
- Posts: 50
- Joined: Fri Dec 02, 2011 2:54 am
Re: Necessity of Driver stage for KT120
The "need" for a driver stage comes from the KT120s requirement for low control grid resistance. This is more of a concern in fixed bias - the KT120 datasheet specifies a maximum G1 resistance of only 51k.
This isn't really because of interelectrode capacitance - the KT120 datasheet lists the input capacitance as 29 pF - but because of a need for a leakage path for electrons that wind up striking the grid. Exceed this value and you greatly increase the likelihood of the tube undergoing thermal runaway.
However in cathode bias, the datasheet allows Rg1 up to 240k, so as long as your grid stopper and grid leak add up to less than that you should be fine.
The problem with requiring a small Rg1 on a power tube - and why so many amp designers routinely ignore the datasheet values - is because the grid leak resistor sets the input impedance of the first stage. Like all coupled stages, there will effectively be a voltage divider formed, comprised of the output impedance of the previous stage into the input impedance of the following stage. So for maximum signal transfer, obviously you want to lower the output impedance of the last stage of the preamp while maximizing the input impedance of the power tube.
The express only uses three triodes for the preamp, so with that spare triode you could either make a regular gain stage using a triode with a lower plate resistance (and thus lower output impedance), or you could use a regular 12AX7 and set the extra stage up as a cathode follower. If you don't want the compression from the typical DC-coupled cathode follower, you can use an AC coupled one. Either way a big grid stopper would be a good idea...
Or, heck, you can just try it as-is, as long as your grid leak and grid stopper combined total less than 240k. Considering the KT120 is a big expensive tube and the datasheet is equal parts pipe dream and photocopied nightmare, I'd personally keep it under 200k total.
This isn't really because of interelectrode capacitance - the KT120 datasheet lists the input capacitance as 29 pF - but because of a need for a leakage path for electrons that wind up striking the grid. Exceed this value and you greatly increase the likelihood of the tube undergoing thermal runaway.
However in cathode bias, the datasheet allows Rg1 up to 240k, so as long as your grid stopper and grid leak add up to less than that you should be fine.
The problem with requiring a small Rg1 on a power tube - and why so many amp designers routinely ignore the datasheet values - is because the grid leak resistor sets the input impedance of the first stage. Like all coupled stages, there will effectively be a voltage divider formed, comprised of the output impedance of the previous stage into the input impedance of the following stage. So for maximum signal transfer, obviously you want to lower the output impedance of the last stage of the preamp while maximizing the input impedance of the power tube.
The express only uses three triodes for the preamp, so with that spare triode you could either make a regular gain stage using a triode with a lower plate resistance (and thus lower output impedance), or you could use a regular 12AX7 and set the extra stage up as a cathode follower. If you don't want the compression from the typical DC-coupled cathode follower, you can use an AC coupled one. Either way a big grid stopper would be a good idea...
Or, heck, you can just try it as-is, as long as your grid leak and grid stopper combined total less than 240k. Considering the KT120 is a big expensive tube and the datasheet is equal parts pipe dream and photocopied nightmare, I'd personally keep it under 200k total.
-
Andy Le Blanc
- Posts: 2582
- Joined: Sat Dec 22, 2007 1:16 am
- Location: central Maine
Re: Necessity of Driver stage for KT120
I've used kt88/90/120 in several builds for guitar, regular old guitar amp circuits were adequate.
Remember that you can use the RC tables in your tube manuals as a rule of thumb
for a circuit design, voltage and operating point to optimize a limited number of tube stages for desired gain.
Get the rig to break up about where you want it.
Remember that you can use the RC tables in your tube manuals as a rule of thumb
for a circuit design, voltage and operating point to optimize a limited number of tube stages for desired gain.
Get the rig to break up about where you want it.
lazymaryamps
Re: Necessity of Driver stage for KT120
Thanks for the replies everyone. Its been a while since I've done any designing for tube electronics so forgive me I'm pretty rusty at this point.
So from what I can tell its more of an impedance issue than one of input sensitivity. At this point I'm thinking something like 56k for the grid stopper and maybe 100k for the grid leak? Or would that be too big?
I might try things as is with those values to see what I can get out of it. Currently its setup with the first two 12ax7s from the express and basically the single ended lead power amp from ax84. It has an extra hole for another noval tube socket but I'd have to do some rearranging on the board layout to make room for the components for another tube and I'd like to avoid that if possible. But I'm in no rush at the moment, the OT should be here Friday and I still need to order the PT, just don't have the funds for it at the moment.
For the person that asked the OT is an Edcor 25w with a 3k primary and 8 ohm secondary:
https://www.edcorusa.com/xse25-8-3k
I called and asked what the primary current rating was and she said 150ma which should be more than enough.
For the PT I'm thinking something like 275-0-275 @ 250ma for the HT and a 5 amp 6.3v heater winding should be plenty for this.
Also as a side note what would be a good screen resistor value for this tube?
Edit: I feel stupid. When you said three triodes I was thinking three preamp tubes. I totally forgot about the extra one that the trainwreck doesn't use. Its been years since Ive done anything with this amp and I totally forgot I made use of that last stage. I'll have to check and see what I did with it as I've long since lost the layout and schematics I drew up for it.
So from what I can tell its more of an impedance issue than one of input sensitivity. At this point I'm thinking something like 56k for the grid stopper and maybe 100k for the grid leak? Or would that be too big?
I might try things as is with those values to see what I can get out of it. Currently its setup with the first two 12ax7s from the express and basically the single ended lead power amp from ax84. It has an extra hole for another noval tube socket but I'd have to do some rearranging on the board layout to make room for the components for another tube and I'd like to avoid that if possible. But I'm in no rush at the moment, the OT should be here Friday and I still need to order the PT, just don't have the funds for it at the moment.
For the person that asked the OT is an Edcor 25w with a 3k primary and 8 ohm secondary:
https://www.edcorusa.com/xse25-8-3k
I called and asked what the primary current rating was and she said 150ma which should be more than enough.
For the PT I'm thinking something like 275-0-275 @ 250ma for the HT and a 5 amp 6.3v heater winding should be plenty for this.
Also as a side note what would be a good screen resistor value for this tube?
Edit: I feel stupid. When you said three triodes I was thinking three preamp tubes. I totally forgot about the extra one that the trainwreck doesn't use. Its been years since Ive done anything with this amp and I totally forgot I made use of that last stage. I'll have to check and see what I did with it as I've long since lost the layout and schematics I drew up for it.
Re: Necessity of Driver stage for KT120
Okay so I finally got around to figuring out what is going on in the second 12ax7, the first triode is a normal gain stage with a 10k cathode resistor, the second appears to be a typical AC coupled cathode follower, except the output is taken from the anode through a coupling cap. Has anyone ever seen a configuration like this before? I'm thinking I probably made a mistake when I built the thing a while back and never caught it, but if not I'm curious to know what it's called.