My first amp build… question
Question:
Hello, I am about to embark on my first tube amp build… Since I already own a beloved 1965 Vox AC10 I want to use that as a standard for voicing my homebrew amp. It is also a very simple circuit to start with, not much more complicated than a Champ. So I will be using the AC10 circuit with two major changes: 1) I will eliminate the vibrato channel all together EF86 and ECF82… and 2) I will be replacing the ECF82 triode and the EF86 input preamps with a single 12AX7 to provide 2 isolated triode inputs. (Since the ECF82 and EF86 will be gone). Tubes will be 12AX7, 12AX7, dual EL84 and EZ81/6CA4 rectifier. I want to keep the tube rectifier because its voltage sag gives the amp a very nice "spongy, boingy" kind of attack, one of the main aspects I like about it. My question is this, would constructing the power supply, preamp, and power amp on 3 separate chassis be of any benefit noise-wise? Or how about just constructing the power supply on a separate chassis? I will be using proper star grounding within each chassis and would connect the various chassis I/O together with shielded cable, etc. I am building this amp primarily for mic’ed recording purposes and want to get the quiescent noise floor as low as possible. i’ve seen some Ampeg amps that use a separate chassis for various stages, was this done primarily for noise or modularity? Any other words of advice before I embark upon this lonely journey would also be greatly appreciated. (BTW I have a deep history of building electronic things, but this is my first venture into a tube project from scratch, but I have built tube Heathkits) I’ve already read all the safety rules here and will be buying a Variac to be even safer, etc.
Response:
– Hide quoted text — Show quoted text -> Hello, > I am about to embark on my first tube amp build… Since I already own > a beloved 1965 Vox AC10 I want to use that as a standard for voicing my > homebrew amp. It is also a very simple circuit to start with, not much > more complicated than a Champ. So I will be using the AC10 circuit > with two major changes: 1) I will eliminate the vibrato channel all > together EF86 and ECF82… and 2) I will be replacing the ECF82 triode > and the EF86 input preamps with a single 12AX7 to provide 2 isolated > triode inputs. (Since the ECF82 and EF86 will be gone). Tubes will be > 12AX7, 12AX7, dual EL84 and EZ81/6CA4 rectifier. I want to keep the > tube rectifier because its voltage sag gives the amp a very nice > "spongy, boingy" kind of attack, one of the main aspects I like about > it. > My question is this, would constructing the power supply, preamp, and > power amp on 3 separate chassis be of any benefit noise-wise?
Yup but hardly worth the effort. > Or how about just constructing the power supply on a separate chassis?
Again – no reason to do that. IMO layout, judicious use of screening and avoiding earth loops is all that’s needed. Some examples: big noise killer: Using screened cable for the input earthed at one end only. Magnetic shielding between transformers, input circuit and guitarist. A properly shielded amp won’t feed back into the pickups at full tilt. Typical good layout: keeping adjacent transformers at 90 degrees to each other, near power stage and well away from input stage. Use cheap plastic pots (seriously) – great for avoiding earth loops 
since it’s your own project you could spare no expense but I’ve cheap plastics that have been working since 1983 without getting scratchy or breaking. Most of that’s dependent on the size and shape of your chassis. For most homebrews, that’s the limiting factor in the budget because one off’s tend to be a little costly. Design the chassis so it sits on the transformers and protects the valves for upside down surgery – that alone reduces hastle;-) Enjoy the ride but don’t expect AGA to be much about amps. cb
Response:
> > power amp on 3 separate chassis be of any benefit noise-wise? > Yup but hardly worth the effort. > Or how about just constructing the power supply on a separate chassis? > Again – no reason to do that. > cb > remainder omitted for brevity…
Thanks Chris, I suspected that there is no magic bullet, just good construction. I was planning on using Aiken Amps 2 point grounding in this article: http://www.aikenamps.com/StarGround.html The first ground star will be at the power supply and cover the PS and power output stage and the second star will cover the input stage preamp. I was not going to depend on the chassis itself. I also planned on keeping all the jacks lifted from the chassis and grounded at one end only to the second star which then connects to the first star in the PS/Output area that then connects to the mains ground which attaches to the chassis at entry. The speaker jacks will also be isolated from the chassis and ferrule run back to the OPT center tap directly. This amp does not use global negative feedback so I dont need to run any other wires from the output jacks back to the preamp area. As for the pots I will ground the pot terminal with a wire directly to star 2 and not connect the grounded pot terminal to the pot case since that will cause a chassis loop, instad the pot case will be lock-washered to the chassis to provide a shield, but the case itself will not touch the ground of the second star. I’m hoping this should eliminate any alternate paths to ground where there might be heavier currents flowing. I forgot about the transformer positions and will do that too. Since this is a one off I most likely will go ahead do the power supply on its own chassis and connect that chassis star ground directly to the pre/power amp chassis star ground, and then use a second wire to ground chassis-to-chassis. This way if the project fails then at least I’ll have a working power supply that I can re-use on a plan B amp. I’m also going to ground my filament legs through 2 star grounded 100 ohm resistors, which was not on the original AC10 schematic but recommended elsewhere here. I’ll also also give the rectifier tube its own winding since the tranny I found has it available anyway. I was going to make a DC filament, but there seems to be some dispute if this is worth it. thanks again, Rick
Response:
Hi Rick, Welcome to AGA. There is a lot of noise here, but I think you will get some help. There are some real experts here. I’ve built a couple and guys were a big help. What you describe tube wise, I built a 5E3 Deluxe variant with the 6CA4, 12AY7, 12AX7, two 6V6GT — very similar to what you propose. I did that because the there was no 5 volt winding available and I had the tube. But as a matter of practicality, you might want to think about using a diode rectifier (full or half, depends on the tranny). Diodes are under 10 cents each, so if you don’t like it, you can always "upgrade". You could even mount the socket and use it to mount the ss rectifier. There is no practical reason I can think of to use more than one chassis and I think there could be some potential headaches. Dismiss this thought. Aiken’s paper on grounds is excellent reading, but you may find making the leap from theory to practice a bit cumbersome. Basically, you run a separate star for each filter cap. Whatever is on that cap — that’s a section. When you get the that point, ask for more advice. On the filament winding, if there is a center tap, you can loose the 100 ohm resistor. If not, that’s the way to go. I have five amps, all with this method. Forget about DC on the filaments. To solder to the chassis, you will need a substantial iron or gun; not the one you use for the rest of the amp. I don’t have one, so I use a torch to solder a solid wire to the chassis, and then anchor to the wire. It’s best to do this before you put anything else on the chassis ;~} Use insulated pots and jacks. You’ll be happy you did. Lead dress is very important. Do take care in the layout planning so there are no long wire runs. Lay DC signal wires close to the chassis, and AC (filament; output tranny) in the air. Twist AC pairs. When DC crosses, do it at a right angle. You shouldn’t need shielded cable. I’ve used my share to compensate for bad layout, though and it’s a learning process. When you get into it, let us know how it’s going. Regards, Phil
– Hide quoted text — Show quoted text -> > power amp on 3 separate chassis be of any benefit noise-wise? > Yup but hardly worth the effort. > > Or how about just constructing the power supply on a separate chassis? > Again – no reason to do that. > cb > remainder omitted for brevity… > Thanks Chris, > I suspected that there is no magic bullet, just good construction. I > was planning on using Aiken Amps 2 point grounding in this article: > http://www.aikenamps.com/StarGround.html > The first ground star will be at the power supply and cover the PS and > power output stage and the second star will cover the input stage > preamp. I was not going to depend on the chassis itself. I also > planned on keeping all the jacks lifted from the chassis and grounded > at one end only to the second star which then connects to the first > star in the PS/Output area that then connects to the mains ground which > attaches to the chassis at entry. The speaker jacks will also be > isolated from the chassis and ferrule run back to the OPT center tap > directly. This amp does not use global negative feedback so I dont > need to run any other wires from the output jacks back to the preamp > area. As for the pots I will ground the pot terminal with a wire > directly to star 2 and not connect the grounded pot terminal to the pot > case since that will cause a chassis loop, instad the pot case will be > lock-washered to the chassis to provide a shield, but the case itself > will not touch the ground of the second star. I’m hoping this should > eliminate any alternate paths to ground where there might be heavier > currents flowing. I forgot about the transformer positions and will do > that too. > Since this is a one off I most likely will go ahead do the power supply > on its own chassis and connect that chassis star ground directly to the > pre/power amp chassis star ground, and then use a second wire to ground > chassis-to-chassis. This way if the project fails then at least I’ll > have a working power supply that I can re-use on a plan B amp. > I’m also going to ground my filament legs through 2 star grounded 100 > ohm resistors, which was not on the original AC10 schematic but > recommended elsewhere here. I’ll also also give the rectifier tube its > own winding since the tranny I found has it available anyway. I was > going to make a DC filament, but there seems to be some dispute if this > is worth it. > thanks again, > Rick
Response:
> Hi Rick, > Welcome to AGA. There is a lot of noise here, but I think you will get some > help. There are some real experts here. I’ve built a couple and guys were > a big help. > the rest removed for brevity…
Thanks Phil, I will check back. I thought about using diodes but realized one of the things I like the most about the AC10 is the "boing factor" that occurs because of the voltage sag against the rectifier on note attacks. I play jazz guitar with an archtop and this bounced yet softer attack is just the thing to reproduce that 1950’s jazz tone of Charlie Christian, Wes Montgomery, Tal Farlow, etc. I may even have to reduce my power transformer size if I cant get this sag to occur again after I am done. I’ve found that amp modelers might get the harmonic content of the waveforms correct, but they can never get the responsiveness and playability correct, ie. something like "boing". For this rason tubes will ive forever at least for guitar players. Still going back and forth on the overall layout though. thanks Rick
Response:
– Hide quoted text — Show quoted text -> > power amp on 3 separate chassis be of any benefit noise-wise? > Yup but hardly worth the effort. > > Or how about just constructing the power supply on a separate chassis? > Again – no reason to do that. > cb > remainder omitted for brevity… >Thanks Chris, >I suspected that there is no magic bullet, just good construction. I >was planning on using Aiken Amps 2 point grounding in this article: >http://www.aikenamps.com/StarGround.html >The first ground star will be at the power supply and cover the PS and >power output stage and the second star will cover the input stage >preamp. I was not going to depend on the chassis itself. I also >planned on keeping all the jacks lifted from the chassis and grounded >at one end only to the second star which then connects to the first >star in the PS/Output area that then connects to the mains ground which >attaches to the chassis at entry.
All these schemes work or don’t work, depending on the exact amp… don’t rely too much on these advice articles… be prepared to fly ground wires around to see where then make the least noise… I use only one ground point in my amps, and they are dead silent. > The speaker jacks will also be >isolated from the chassis and ferrule run back to the OPT center tap >directly.
Spk jacks don’t matter if there is no feedback… > This amp does not use global negative feedback so I dont >need to run any other wires from the output jacks back to the preamp >area. As for the pots I will ground the pot terminal with a wire >directly to star 2
Want some advice? Run the ground to the next stage grid resistor ground point, through shield cable if you can… Running a ground wire from the middle of the tone/volume stack out to the star point could make a ground loop… you may want to prepare to ‘fly’ this wire around… connect where it is quietest. > and not connect the grounded pot terminal to the pot >case since that will cause a chassis loop, instad the pot case will be >lock-washered to the chassis to provide a shield, but the case itself >will not touch the ground of the second star. I’m hoping this should >eliminate any alternate paths to ground where there might be heavier >currents flowing. I forgot about the transformer positions and will do >that too. >Since this is a one off I most likely will go ahead do the power supply >on its own chassis and connect that chassis star ground directly to the >pre/power amp chassis star ground, and then use a second wire to ground >chassis-to-chassis. This way if the project fails then at least I’ll >have a working power supply that I can re-use on a plan B amp.
More advice… bond the chassis’s together with LARGE copper braid… >I’m also going to ground my filament legs through 2 star grounded 100 >ohm resistors,
You can ground the filament anywhere on the chassis, sometimes running it to the star ADD’s noise by injecting AC directly into the star point. >which was not on the original AC10 schematic but >recommended elsewhere here. I’ll also also give the rectifier tube its >own winding since the tranny I found has it available anyway. I was >going to make a DC filament, but there seems to be some dispute if this >is worth it.
not worth shitola… >thanks again, >Rick
Bob amp builder. http://www.geocities.com/flintstudio/index.html
Response:
Bob, I checked out your web site, and I see you’ve been busy. I especially liked the prototype #10 that single ended EL84 with 12AX7. I know its only 5 watts but I think a lot of jazz and acoustic players would be interested in a very small tube head under 15 pounds that can be housed in a shoulder-strapped gig bag. Add a small separate speaker cabinet loaded with a 103db/1w 12 incher and a switchable piezo, put an extendable suitcase handle and wheels on it and you’re set. A person can carry the guitar, the head, and roll the speaker all in one trip from the car. I say a separate speaker and head because with acoustics and archtops you need to get the speaker father away from you to avoid feedback, but players still like to have the head controls at hands reach. A setup similar to that little new "Crate Powerblock" thats out now, except with tubes. With a high efficiency speaker those 5 watts can go a long way. Nice stuff. thanks Rick
Response:
– Hide quoted text — Show quoted text -> > power amp on 3 separate chassis be of any benefit noise-wise? > Yup but hardly worth the effort. > > Or how about just constructing the power supply on a separate chassis? > Again – no reason to do that. > cb > remainder omitted for brevity… > Thanks Chris, > I suspected that there is no magic bullet, just good construction. I > was planning on using Aiken Amps 2 point grounding in this article: > http://www.aikenamps.com/StarGround.html > The first ground star will be at the power supply and cover the PS and > power output stage and the second star will cover the input stage > preamp. I was not going to depend on the chassis itself. I also > planned on keeping all the jacks lifted from the chassis and grounded > at one end only to the second star which then connects to the first > star in the PS/Output area that then connects to the mains ground which > attaches to the chassis at entry. The speaker jacks will also be > isolated from the chassis and ferrule run back to the OPT center tap > directly. This amp does not use global negative feedback so I dont > need to run any other wires from the output jacks back to the preamp > area. As for the pots I will ground the pot terminal with a wire > directly to star 2 and not connect the grounded pot terminal to the pot > case since that will cause a chassis loop, instad the pot case will be > lock-washered to the chassis to provide a shield, but the case itself > will not touch the ground of the second star. I’m hoping this should > eliminate any alternate paths to ground where there might be heavier > currents flowing. I forgot about the transformer positions and will do > that too. > Since this is a one off I most likely will go ahead do the power supply > on its own chassis and connect that chassis star ground directly to the > pre/power amp chassis star ground, and then use a second wire to ground > chassis-to-chassis. This way if the project fails then at least I’ll > have a working power supply that I can re-use on a plan B amp. > I’m also going to ground my filament legs through 2 star grounded 100 > ohm resistors, which was not on the original AC10 schematic but > recommended elsewhere here. I’ll also also give the rectifier tube its > own winding since the tranny I found has it available anyway. I was > going to make a DC filament, but there seems to be some dispute if this > is worth it. > thanks again, > Rick
What Chris said plus: – use a PT with a filament power ground. They are quieter than the ones that don’t have a ground and you have to use resistors to make one – consider rectifying the pre-amp tube filament power to run the pre- off of smoothed and filtered DC. A hassle but can be worth the effort in a recording studio type low output amp – use as good a Iron (PT and OT) as you can get/afford – since you’re using a star ground system, use a heavy aluminum chassis, not the flimsy hobby-type hammond thin stuff – sprague filter caps – I’d probably change the rectifier to a 5Y3 or 5V4 instead of the 6CA4, but that’s just ’cause I know the pinouts without having to look ‘em up <G> Good luck and keep us posted. Walt Campbell Campbell Sound http://www.campbellsound.com/
Response:
>Bob, >I checked out your web site, and I see you’ve been busy. I especially >liked the prototype #10 that single ended EL84 with 12AX7. I know its >only 5 watts but I think a lot of jazz and acoustic players would be >interested in a very small tube head under 15 pounds that can be housed >in a shoulder-strapped gig bag. Add a small separate speaker cabinet >loaded with a 103db/1w 12 incher and a switchable piezo, put an >extendable suitcase handle and wheels on it and you’re set. A person >can carry the guitar, the head, and roll the speaker all in one trip >from the car.
I’m thinking of making that amp in such a way as it can be detached from the speaker box and used as a head… or clamped onto any box… >I say a separate speaker and head because with acoustics and archtops >you need to get the speaker father away from you to avoid feedback, but >players still like to have the head controls at hands reach. A setup >similar to that little new "Crate Powerblock" thats out now, except >with tubes. With a high efficiency speaker those 5 watts can go a long >way. >Nice stuff. >thanks >Rick
Glad you like my stuff!
Response:
>I play jazz guitar with an archtop and this bounced yet > softer attack is just the thing to reproduce that 1950’s jazz tone of > Charlie Christian, Wes Montgomery, Tal Farlow, etc.
Not, Rick. CC stopped playing & recording in 1941, by the way. NO ONE has ever been able to nail his tone – even Pass bemoaned his feeble attempts to do so late in his career, and his tone is still the benchmark. He did not play through anything like what you are talking about. He played through a very early and crude tube amp of about 10 watts with an electromagnetic speaker that had no "sag" whatsoever. These were both choke-input and class A designs no one builds today. You have been misinformed. You will not sounds anything like CC with any kind of amp that has poor B+ regulation (sag) like popular guitar amps do, and it will strongly work against you. CC played with ALL DOWNSTROKES through amps that held the full but lower initial power of each note, instead of the initial punch & sag now widely used. Having the input choke also be the speaker magnet further adds a whole other world to this tonally which is unobtainable from anything else. If you want to sound remotely like him, you must have a 12" no-breakup EM spkr in a zero-sag amp with enough power to have the speaker’s dynamics and the solid, no-punches-pulled evenness of delivery & performance. Then, you must learn to play with mostly downstrokes and with authority. The idea of getting anything like this from a smaller and more versatile, albeit very nice, little amp of modern conventional design with a PM spkr, is 180 degrees away from the goal. Save yourself dubbing around for 40 years proving it, as I did. NONE of this is any slight to Bob, who is a conscientious fellow builder very interested in his customer’s success, and no dummy. All of us are pursuing the low-power tonal/cost/package with the best excellence we can bring to it, but this is a different horse in a different barn.
Response:
– Hide quoted text — Show quoted text ->I play jazz guitar with an archtop and this bounced yet > softer attack is just the thing to reproduce that 1950’s jazz tone of > Charlie Christian, Wes Montgomery, Tal Farlow, etc. >Not, Rick. >CC stopped playing & recording in 1941, by the way. NO ONE has ever >been able to nail his tone – even Pass bemoaned his feeble attempts to >do so late in his career, and his tone is still the benchmark. He did >not play through anything like what you are talking about. He played >through a very early and crude tube amp of about 10 watts with an >electromagnetic speaker that had no "sag" whatsoever. These were both >choke-input and class A designs no one builds today. You have been >misinformed. You will not sounds anything like CC with any kind of amp >that has poor B+ regulation (sag) like popular guitar amps do, and it >will strongly work against you. >CC played with ALL DOWNSTROKES through amps that held the full but >lower initial power of each note, instead of the initial punch & sag >now widely used. Having the input choke also be the speaker magnet >further adds a whole other world to this tonally which is unobtainable >from anything else. >If you want to sound remotely like him, you must have a 12" no-breakup >EM spkr in a zero-sag amp with enough power to have the speaker’s >dynamics and the solid, no-punches-pulled evenness of delivery & >performance. Then, you must learn to play with mostly downstrokes and >with authority. >The idea of getting anything like this from a smaller and more >versatile, albeit very nice, little amp of modern conventional design >with a PM spkr, is 180 degrees away from the goal. Save yourself >dubbing around for 40 years proving it, as I did. >NONE of this is any slight to Bob, who is a conscientious fellow >builder very interested in his customer’s success, and no dummy. All >of us are pursuing the low-power tonal/cost/package with the best >excellence we can bring to it, but this is a different horse in a >different barn.
I have an old radio here with an electromagnet speaker, the owner never came back for it… I should try building it into an amp! I wonder what kind of power it needs to draw… I also wonder if any of those old amps had weirder circuits like triode power out, or pentode preamps…
Response:
Related Posts