Recently I've been toying with the idea of a really low output tube amp. I've done a couple Bogen conversions that have turned out really nice (especially the CHB-10A) but now I want to do something from scratch.
I came across this site a couple days ago and after doing a bit more research I think this is what I'll be doing hopefully in the next couple weeks.
http://www.sopht.ca/mambo/index.php?option=com_content&task=view&id=25&Itemid=37
I plan on building something similar to Version 2 of the 12AL8 amp. Aside from the unbypassed cathodes (does this really make a difference?) it all seems to look alright, and I'm impressed with the sound clips posted there. I really like the idea of using a 12V supply for everything. Working with high voltages doesn't really bother me, but it'd be nice to be able to just plug in a wall wart and be done with it rather that go through the trouble of designing and building a new power supply.
That project should go fairly smoothly. I'll probably build it without an enclosure on my workbench first so I can swap stuff out, then rebuild it in an enclosure. The one question I have though is what I should use for an output transformer. The amp he built has a Hammond 119DA, which is really a speaker matching transformer. It seems like it would work alright, but I can't find anywhere that has them in stock. Any other ideas for an OPT? Will it be a problem if I use, say, a 1000:8 ohm general purpose audio output tansformer from radio shack? With the low wattage it seems like this could be an option, but would it be a problem?
Eventually I want to build one to use as an overdrive pedal like the site mentions. My question here also has to do with impedances. If I use it as a pedal, should I use an output transformer at all, or should I just wire it like a preamp tube and hook the output straight to a 1/4" jack?
Thanks in advance.
-Darren
I came across this site a couple days ago and after doing a bit more research I think this is what I'll be doing hopefully in the next couple weeks.
http://www.sopht.ca/mambo/index.php?option=com_content&task=view&id=25&Itemid=37
I plan on building something similar to Version 2 of the 12AL8 amp. Aside from the unbypassed cathodes (does this really make a difference?) it all seems to look alright, and I'm impressed with the sound clips posted there. I really like the idea of using a 12V supply for everything. Working with high voltages doesn't really bother me, but it'd be nice to be able to just plug in a wall wart and be done with it rather that go through the trouble of designing and building a new power supply.
That project should go fairly smoothly. I'll probably build it without an enclosure on my workbench first so I can swap stuff out, then rebuild it in an enclosure. The one question I have though is what I should use for an output transformer. The amp he built has a Hammond 119DA, which is really a speaker matching transformer. It seems like it would work alright, but I can't find anywhere that has them in stock. Any other ideas for an OPT? Will it be a problem if I use, say, a 1000:8 ohm general purpose audio output tansformer from radio shack? With the low wattage it seems like this could be an option, but would it be a problem?
Eventually I want to build one to use as an overdrive pedal like the site mentions. My question here also has to do with impedances. If I use it as a pedal, should I use an output transformer at all, or should I just wire it like a preamp tube and hook the output straight to a 1/4" jack?
Thanks in advance.
-Darren
Try the 1000: 8 transformer - it's not TOO far off... but I think you'll find that diferent transformers give different sounds. 600 : 8 shouldn't be too hard to find. Even for a pedal, I think the transformer will be best. You may have to attenuate the output a bit though...
As a pedal... It'd be cool to go ahead and use the OT connected to an 8 ohm resistive load, then using a voltage divider off of that to create a line out... That way you can get roughly the overdriven tone of the little amp into the front of the big amp and power scale it. You'd also be able to add another jack as a speaker out to bypass the resistive load.
And a Fender replacement Reverb Transformer would be a suitable Output transformer as well. I've used it with a paralleled 12AT7 as a single ended output tube and it soundedpretty darned good. (After all the send of a Fender reverb circuit is indeed a small single ended power amp.)
And a Fender replacement Reverb Transformer would be a suitable Output transformer as well. I've used it with a paralleled 12AT7 as a single ended output tube and it soundedpretty darned good. (After all the send of a Fender reverb circuit is indeed a small single ended power amp.)
I was just about to ask about a voltage divider before I saw that post 🙂. Thanks. I'm thinking of swapping a 50k or 100k pot in place of one of the resistors as kind of an output volume control. Both my amps are set up with a grid-leak biased first preamp stage so they don't take too well to a large input signal.
The reverb driver transformers are somewhere in the neigborhood of 15k:8 ohms. It matches nice to the plate resistance of a 12AT7, but looks way too high for the 12AL8.
I was doing a bit of research and I'm a little confused now.
Here's the Sylvania data sheet I've been using:
http://frank.pocnet.net/sheets/137/1/12AL8.pdf
It claims that the plate resistance is only 480 ohms for class A use, but there's another category listed as "Resistance-coupled cascaded amplifier" and it lists 800 ohms as the load resistance for the tetrode section. This is what I was going by, is this really what I should be looking at when choosing a transformer?
I also just found this:
http://www.nj7p.org/Tube4.php?tube=12al8
This claims the plate resistance is about 1k for a class A amp and says nothing about a resistance-coupled use. Does this mean a 1000:8 transformer would work the best?
Sorry about the stupid questions, I've never had to deal with choosing a transformer for any of my other projects. Thanks.
-Darren
The reverb driver transformers are somewhere in the neigborhood of 15k:8 ohms. It matches nice to the plate resistance of a 12AT7, but looks way too high for the 12AL8.
I was doing a bit of research and I'm a little confused now.
Here's the Sylvania data sheet I've been using:
http://frank.pocnet.net/sheets/137/1/12AL8.pdf
It claims that the plate resistance is only 480 ohms for class A use, but there's another category listed as "Resistance-coupled cascaded amplifier" and it lists 800 ohms as the load resistance for the tetrode section. This is what I was going by, is this really what I should be looking at when choosing a transformer?
I also just found this:
http://www.nj7p.org/Tube4.php?tube=12al8
This claims the plate resistance is about 1k for a class A amp and says nothing about a resistance-coupled use. Does this mean a 1000:8 transformer would work the best?
Sorry about the stupid questions, I've never had to deal with choosing a transformer for any of my other projects. Thanks.
-Darren
Ah... I thought you'd be using the 12U7... which would be pretty decent with that transformer... You are on the right path...
So, now I'm thinking of grabbing a cheap 70V line transformer as well. These have several primary taps, but they're labelled by wattage instead of impedance.
I've been trying to figure out the impedances using ohm's law, someone tell me if I'm on the right track.
His site claims that the 2.5 watt tap is best on the 70V line transformer. 2.5 watts on the 8 ohm tap comes to about 4.5V if my calculations are right. Turns ratio would be 70:4.5 or about 15.5:1. Square it to get the impedance ratio and you get a little over 240:1 or right around 2000:8, not optimal at all. Did I do that right? If that's the case, the 5 watt tap gives me 1000:8, and the 10 watt tap gives me 500:8. Still plenty to play with.
I've been trying to figure out the impedances using ohm's law, someone tell me if I'm on the right track.
His site claims that the 2.5 watt tap is best on the 70V line transformer. 2.5 watts on the 8 ohm tap comes to about 4.5V if my calculations are right. Turns ratio would be 70:4.5 or about 15.5:1. Square it to get the impedance ratio and you get a little over 240:1 or right around 2000:8, not optimal at all. Did I do that right? If that's the case, the 5 watt tap gives me 1000:8, and the 10 watt tap gives me 500:8. Still plenty to play with.
Well, I got it all wired up pretty much exactly like the 12AL8 amp, minus the 200 ohm resistor between the tetrode and the OPT. I ended up using a line matching transformer.
The volume's decent into a speaker, but the amp doesn't sound too good. I checked between all the stages with my oscilloscope, and found something a bit odd. I've used a scope long enough to know it wasn't good, but I'm not sure what it means. I noticed that after each stage, the "lower" half of the wave was much larger than the upper part. This was most noticeable between the second and third stage, so I'm pretty sure it's the way the 12U7 is wired. I wish I could get a picture but trying to take a picture of a CRT scope and getting anything legible isn't really possible 🙂. There was a huge difference between the top and bottom halves of the wave though. At the output, it's pretty obvious that the bottom part is clipping pretty hard, especially with chords stummed hard, but the top really isn't.
Any ideas? Should I try cathode-biasing the first preamp stages?
-Darren
The volume's decent into a speaker, but the amp doesn't sound too good. I checked between all the stages with my oscilloscope, and found something a bit odd. I've used a scope long enough to know it wasn't good, but I'm not sure what it means. I noticed that after each stage, the "lower" half of the wave was much larger than the upper part. This was most noticeable between the second and third stage, so I'm pretty sure it's the way the 12U7 is wired. I wish I could get a picture but trying to take a picture of a CRT scope and getting anything legible isn't really possible 🙂. There was a huge difference between the top and bottom halves of the wave though. At the output, it's pretty obvious that the bottom part is clipping pretty hard, especially with chords stummed hard, but the top really isn't.
Any ideas? Should I try cathode-biasing the first preamp stages?
-Darren
Please, someone on here has to know what I'm talking about.
I tried cathode-biasing the first preamp stages with no improvement. I used a pot in place of the cathode resistor and tweaking it only yielded different volumes, it was still uneven.
This distortion sounds a lot like the distortion you get when you keep playing through an amp for a few seconds after you shut it off. Very harsh, and unpleasant.
I'm going to try some larger plate resistors. Anyone have any other ideas? Please? I wanted to get this working good beore tomorrow.
I tried cathode-biasing the first preamp stages with no improvement. I used a pot in place of the cathode resistor and tweaking it only yielded different volumes, it was still uneven.
This distortion sounds a lot like the distortion you get when you keep playing through an amp for a few seconds after you shut it off. Very harsh, and unpleasant.
I'm going to try some larger plate resistors. Anyone have any other ideas? Please? I wanted to get this working good beore tomorrow.
Increasing the first stage's plate resistor got me a better sound overall, but I'm still getting the strange unbalanced wave between the stages.
I can get sound cool sounds by running the output straight into my CHB-10A. Hopefully I can get this figured out.
I can get sound cool sounds by running the output straight into my CHB-10A. Hopefully I can get this figured out.
I have to admit that it is a mystery to me, but nobody else is responding, so I'll take a blind stab.
Assuming you don't have your scope set to invert, the behavior suggest that the 12U7 is biased cold. This is not what I would expect with 0 volt bias. It's possible, I suppose, that your grids are drawing current. You could try reducing the grid resistors on the first two stages from 1M to something like 220k or 100k just to see what happens.
If you happen to have another 12U7 you might plug it in to see if there's a difference.
-- Dave
Assuming you don't have your scope set to invert, the behavior suggest that the 12U7 is biased cold. This is not what I would expect with 0 volt bias. It's possible, I suppose, that your grids are drawing current. You could try reducing the grid resistors on the first two stages from 1M to something like 220k or 100k just to see what happens.
If you happen to have another 12U7 you might plug it in to see if there's a difference.
-- Dave
Thanks for the reply.
Yep, another 12U7 was the first thing I tried - no change.
As far as bias goes, I'm getting about -.97VDC on the first stage and about -280mv on the second, which seems like it's running really hot rather than cold.
Right now the first stage is set up almost exactly like the first stage of the 12K5 amp (only difference is a 69k instead of 47k input resistor) and the rest of the amp is set up like version 2 of the 12AL8 amp. This seems odd because he has oscilloscope pictures of the 12K5 amp and he claims that it's perfectly symmetrical. I've run a 1khz sine wave into it and although on the first stage it looks symmetrical, a vast majority of the wave still dips well below the baseline.
The ideal bias voltage is about -.75, so even though the first stage is a bit cold, I don't think that can result in the huge difference in the top and bottom parts of the wave. I'll keep messing with it. Does the value of the bypass cap really make a difference? I'm using a 1.5uF, like the schematic calls for. Should I try something smaller?
Yep, another 12U7 was the first thing I tried - no change.
As far as bias goes, I'm getting about -.97VDC on the first stage and about -280mv on the second, which seems like it's running really hot rather than cold.
Right now the first stage is set up almost exactly like the first stage of the 12K5 amp (only difference is a 69k instead of 47k input resistor) and the rest of the amp is set up like version 2 of the 12AL8 amp. This seems odd because he has oscilloscope pictures of the 12K5 amp and he claims that it's perfectly symmetrical. I've run a 1khz sine wave into it and although on the first stage it looks symmetrical, a vast majority of the wave still dips well below the baseline.
The ideal bias voltage is about -.75, so even though the first stage is a bit cold, I don't think that can result in the huge difference in the top and bottom parts of the wave. I'll keep messing with it. Does the value of the bypass cap really make a difference? I'm using a 1.5uF, like the schematic calls for. Should I try something smaller?
No deal on the grid resistors. It's still bad.
I also biased the first stage to about -.75, and still had problems.
I'm out of ideas. I'll throw a 12AX7 in there for the heck of it (same pinout as the 12U7) and after that I'm probably going to give up. This is very odd.
I also biased the first stage to about -.75, and still had problems.
I'm out of ideas. I'll throw a 12AX7 in there for the heck of it (same pinout as the 12U7) and after that I'm probably going to give up. This is very odd.
No deal with the 12AX7.
I found something that may explain most of it though. I've been hooking the scope to the grid side of the coupling cap, not the plate side. When I do that, it looks a lot better. The top part of the wave is getting clipped and the bottom isn't, but it looks a lot better than I've been seeing. I bet if I rebias both stages it'll end up looking good after all. I guess I've still got a ways to go before I master the scope 🙂.
Tonight I'll cathode bias the second stage and increase the plate resistor, then I'll probably call it done until I can put it in an enclosure.
I found something that may explain most of it though. I've been hooking the scope to the grid side of the coupling cap, not the plate side. When I do that, it looks a lot better. The top part of the wave is getting clipped and the bottom isn't, but it looks a lot better than I've been seeing. I bet if I rebias both stages it'll end up looking good after all. I guess I've still got a ways to go before I master the scope 🙂.
Tonight I'll cathode bias the second stage and increase the plate resistor, then I'll probably call it done until I can put it in an enclosure.
Sounds like you were driving the second stage into grid current, so rebiasing the second stage should help. Note that you will not be able to swing the grid significantly more positive than the cathode as this is the point when grid current starts to flow - looks just like diode clipping. 😀
That makes sense until you think about the phase. He's getting more gain on the negative going part of the cycle; that's when the grid is going positive.... Unless his scope is set to invert and he didn't notice.
-- Dave
-- Dave
- Status
- Not open for further replies.