The Bax circuit has less insertion losses than the standard FMV circuit but by how much depends. I just was thinking of using it because of its prevalence in HIFI but as stated, there may need to be some tailoring to the rest of the amp to make it sound right. It's also a different style where at the 12 o'clock position you're at a neutral or flat response. Turning counterclockwise is subtractive and clockwise is additive. Just was some thoughts I had 😛
Also for PSUD there's a few things to think of. Ripple, and sagging that may result from load, your estimated output voltage, etc.
I had thought about putting a second VVT in because I've got some odd tubes I'm thinking about plugging in. The idea came from the "hundred-buck-amp" challenge they had on here a while back. Rummaging around for oddball noval tubes is going to be fun and I'd like the capability to adjust per what I might come across.
Our (yours and mine) purposes vary from the sounds of it though! For me this build is going to be less of making just an amp and more of making a test mule I can play around with. I obviously want something that sounds good but I might be changing around what I have in it even for just academic purposes 😛
That Adam's site was very helpful, thanks jlangholzj.
I'm now keen to try the 'moonlight' stack.
I think our purposes are actually quite close on this one. My Matchless build was pretty much by the book (I made some very minor mods after the fact). But with this 'champ' I want to experiment and further my understanding of what's happening at each step in the circuit. I don't want to end up with an amp that has 1000 pots and switches crammed onto a front panel - as they just become frustrating in use. The plan is to mount a lot of the adjustment internally or on the back panel, get the amp into the sweet spot for it's cab and speaker (and my ears) and then use the main controls to adjust for different guitar/pedal setups. I will attempt to lay things out internally in a way that facilitates easy re-soldering too.
I'm now keen to try the 'moonlight' stack.
I think our purposes are actually quite close on this one. My Matchless build was pretty much by the book (I made some very minor mods after the fact). But with this 'champ' I want to experiment and further my understanding of what's happening at each step in the circuit. I don't want to end up with an amp that has 1000 pots and switches crammed onto a front panel - as they just become frustrating in use. The plan is to mount a lot of the adjustment internally or on the back panel, get the amp into the sweet spot for it's cab and speaker (and my ears) and then use the main controls to adjust for different guitar/pedal setups. I will attempt to lay things out internally in a way that facilitates easy re-soldering too.
The approach I use is to use the highest value resistor I expect to use for a particular purpose and then I just have to parallel resistors in to lower the value, no re-soldering to speak of. Get a set of alligator jumpers and parallel them. Also with caps, I use the lowest value I expect will sound good and parallel others on to increase the value. Again, no re-soldering needed. Or you could get a decade box. I have a cap/resistor box and it works very well. it's nice to do it this way because you can hear the difference right away so it's easier to compare the different values almost right away, especially with the decade box. Just dial in a value that sounds good, and then figure the total and then solder in that value for the final build.
Last edited:
That makes sense boobtube. The decade boxes are probably out of my price range, but such a good idea... hmmmmmm. I might actually breadboard this one, well, at least the tone stack.
Updated schematic attached, cleaned up, named etc.
Can't figure out what OT to go for. Hammond 16XX are big money. Does anyone have any experience of Edcor GXSE's?
Updated schematic attached, cleaned up, named etc.
Can't figure out what OT to go for. Hammond 16XX are big money. Does anyone have any experience of Edcor GXSE's?
Looks like you updated the name but not the scheme of your tone stack. The moonlight is only a single tone knob with a vol pot. The current one looks like a FMV yet.
I'm also no PSU master....so take this next bit with a grain of salt but you really need to change your capacitor values and change your references. I wouldn't take my reference before the inductor. Less ripple, less noise. That choke's doing some filtering work for you and there's no use in pulling it before then. Also your caps are TINY. make them bigger. Even looking at the original 5F2 they're 450u caps, considerably larger than the 16u that you've got in there.
here's what your current PSU design looks like on 220V secondaries:
Look at that HUGE voltage spike!!! I hear tubes going poof already.....ewps.
Here's what I've got brewed up:
I included the Pre-filter voltage to show how much it really does oscillate on startup as well. I tried to hit the original ~300 and ~260 on the 5F2. These will most likely need to get biased UP after I get the whole circuit in but I was most interested in the transients and what they were doing. Worst case is that I have to bias them up and my tubes will just be a bit cold the first time through...no harm done. Of course this doesn't have the VVT in it either, which would replace what the 3.6k in my CKT is doing and would also allow for changing the plate voltage once load gets applied.
I'm also no PSU master....so take this next bit with a grain of salt but you really need to change your capacitor values and change your references. I wouldn't take my reference before the inductor. Less ripple, less noise. That choke's doing some filtering work for you and there's no use in pulling it before then. Also your caps are TINY. make them bigger. Even looking at the original 5F2 they're 450u caps, considerably larger than the 16u that you've got in there.
here's what your current PSU design looks like on 220V secondaries:
Look at that HUGE voltage spike!!! I hear tubes going poof already.....ewps.
Here's what I've got brewed up:
I included the Pre-filter voltage to show how much it really does oscillate on startup as well. I tried to hit the original ~300 and ~260 on the 5F2. These will most likely need to get biased UP after I get the whole circuit in but I was most interested in the transients and what they were doing. Worst case is that I have to bias them up and my tubes will just be a bit cold the first time through...no harm done. Of course this doesn't have the VVT in it either, which would replace what the 3.6k in my CKT is doing and would also allow for changing the plate voltage once load gets applied.
Last edited:
You are looking at the voltage rating, not the uF rating. The original 5F1 Champ has 16,8,8uF for the nodes. Also, the 5F2 you quote is a Princeton, not a Champ and it had 8,8,8uF for it's nodes. Also, if you study guitar amp schematics, most of them take the plate supply before the choke and the screen after the choke, at least almost all NEWER designs. The 5F1 Champ has a 10k dropping resistor, not a choke. Lightreel has increased the 16uF node to a 33uF and the second and third from 8uF to 16uF, so he had doubled the original circuit's filtering. That should do just fine.
Last edited:
aaannnddd you are correct. I swear I dont need glasses.... Also yes, I'm basing mine off the 5f2 and I think paul is as well....could be wrong though.
So after hooking load up then does it really take that much overshoot out of it? Or are the tubes just that stout that the ~half second of transient spike seen isn't of consequence. I guess I've been staring so much at my 25W hifi lately that I've got 150u values lodged in my head.
EDIT:
going back and looking at this, using the original 5F2 values its...sub tenth of a second on that transient spike. don't think that will be an issue at all. mah bad. something about learning process. Chances are too the estimation I had of the 100k load is a bit high, coming down a decade shows a considerably closer result to the original 5f2 with their values.
Last edited:
The design goals in the guitar amp world are at odds with the hi-fi world. We welcome distortion, although this isn't always the goal as I've heard many others say it is. We like clean sounds too. But our definition of clean is still more distorted than hi-fi clean. Many of the shortcomings of older designs are sometime the tone we are looking for or contribute to them. This is why many of the older designs are still emulated these days, the 5F1 Champ and 5F2 Princeton being two of them.
You're correct. Post #24 says modified 5F2 in his schematic title. My Bad. And yes, the tone stack needs to be updated.
You're correct. Post #24 says modified 5F2 in his schematic title. My Bad. And yes, the tone stack needs to be updated.
Last edited:
Loving all the insights here. Learning "out loud" is the only way I learn!
Sketch must have outputted the wrong layer for the tone stack (I have quite a few!), I'll update again tomorrow.
Any thoughts on Edcor transformers?
Sketch must have outputted the wrong layer for the tone stack (I have quite a few!), I'll update again tomorrow.
Any thoughts on Edcor transformers?
I've heard good things about the edcor's from the guys in the hifi world. Look at the 17xx series hammonds. a 1750E looks like a suitable OPT for the amp and is considerably cheaper than the 16xx counterpart. The downside is that it's got a narrower frequency band than the 16xx counterpart as well.....can't hav eit all i guess. 8 ohm output doesn't bother me too much. GSW will wind the speaker for 4,8, or 16 when you order it and if I end up wanting to run this into a 2x10 or something I can always just wire series/parallel to hit the 8ohm target.
I think the 1750E is push-pull. Did you mean something else?
Hammond only seem to do really cheap and nasty or really expensive but overkill single-ended transformers. Hence my wondering about Edcor (a middle ground?)
Hammond only seem to do really cheap and nasty or really expensive but overkill single-ended transformers. Hence my wondering about Edcor (a middle ground?)
sssshhhhhh I was looking at the 1750E for a different project 😛 You are correct though it's a push-pull. The 1750C is the suitable replacement for the champ.
Truthfully I've got no experience one way or another with either. They're two of the most "recognizable" names off the top of my head. I know several guys have used the edcor's for OPT's on amps though to their liking. Why not buy one and roll with it. Worst case you're out ~$30 tops on a DIY that the purpose is to find out what you do/don't like!
Actually, for a guitar OT, you don't want to have the freq. response much higher than 10kHz. or so. It will just reproduce hiss and other noise more. I can't recall where I read this but I have a few books and one of them is this one: Guitar Amplifier Power Amps Book It takes a while to absorb this one's info as it's geared to pros and engineers but it covers about everything you can imagine for guitar amps power section. It takes into account that they will be overdriven and has a chapter devoted just to this subject. And the trannies for higher freq. response are generally more expensive also.
I've got another one that I'm picking up as well that was recommended. "Valve Amplifiers" by Morgan Jones. Quite a few people have said that they like that one as well.
The one you mentioned looks like lots of fun actually....they never taught us anything about different classes of amps in school I'm afraid. The most we ever got was "here's an op-amp" and the basics of what it does. Fet's were always used as a switch and outside of the small signal model we never did anything with BJT's. So I'm flying by the seat of my pants on this one!
What you're saying on the frequency response makes a lot of sense, given the audio spectrum. Just didn't know if that narrower response mean it would start rolling off around areas I was concerned with. I guess an open E is still up at 300+ Hz and the high F is just under 6k. So even if it were a concern there's plenty of room for leway there.
On the frequency thing, open 6th string (E) on the guitar is 82Hz and 1st string 20th fret (C) is 1047Hz.
A helpful chart
Harmonics are integer multiples of those fundamentals... but even relatively high-order harmonics of guitar notes are still way below 10kHz.
I do find it odd when amps are strongly rolled off below 10kHz however, as the pleasant low-level white noise becomes noticeably absent.
A helpful chart
Harmonics are integer multiples of those fundamentals... but even relatively high-order harmonics of guitar notes are still way below 10kHz.
I do find it odd when amps are strongly rolled off below 10kHz however, as the pleasant low-level white noise becomes noticeably absent.
Also, here's a scheme idea for the DC heater supply. Pretty simple but I figured I'd throw it in here for reference. I don't think that large of a C1 is necessary, it was the topology i was after mostly.
http://i1245.photobucket.com/albums/gg593/Lebowski84/DC_filament_supply_schematic.png
an LM350 should give enough oompf to drive both of them. "Them" being the 12zx7's in parallel and the 6v6 heater. If not then you can always add a second LM350 and drive more heaters with that. I had to do something similar with my old Harmon Kardon when I converted over the heaters. The original 7350's draw 800mA on the heaters! x4 plus another x5 12**7's ....needed more than one 😛
http://i1245.photobucket.com/albums/gg593/Lebowski84/DC_filament_supply_schematic.png
an LM350 should give enough oompf to drive both of them. "Them" being the 12zx7's in parallel and the 6v6 heater. If not then you can always add a second LM350 and drive more heaters with that. I had to do something similar with my old Harmon Kardon when I converted over the heaters. The original 7350's draw 800mA on the heaters! x4 plus another x5 12**7's ....needed more than one 😛
There is one small thing about the coupling cap size to the 6V6. For some reason that I can't recall, SE amps have lower bass response than push pull amps. I can attest to that since I have both. If you are short on bass, increase that .022 cap(C3) to .047 or.1uF to increase bass. Otherwise, I agree the schema looks great. If you have another .022, just parallel it on to see what happens or whatever size sounds good. If it sounds fine as is, tell me to go jump in the lake! C1 is a good value and a standard value if you want to fully bypass that stage, although 10uF would be fine too.
Last edited:
- Status
- Not open for further replies.