Single Ended Layout

[spacelabstudio]

Does this look sane? 

http://spacelabstudio.com/amps/se1/se1_layout.png
http://spacelabstudio.com/amps/se1/se1.txt

Thanks!
Chris

[DummyLoad]

the green font is difficult for my old eyes to read. the heater pins for the 12AX7s are wrong in the schematic  http://spacelabstudio.com/dearmondschematictx1.gif - you could have just posted the schematic? ;)

regards,

--DL

[spacelabstudio]

the green font is difficult for my old eyes to read. the heater pins for the 12AX7s are wrong in the schematic  http://spacelabstudio.com/dearmondschematictx1.gif - you could have just posted the schematic? ;)

Thanks.  I'm mostly hoping for feedback on the layout.  I made the green darker in hopes that it's easier to read with better contrast against the white background.  I didn't noticed they'd gotten the heater pins wrong on the Dearmond schematic.  That's funny! 

Chris

[TIMBO]

Hi Space,I quite enjoy doing the schems and layouts gets the old grey matter stirring.I tend to put way too much stuff on as info but i find that it may help down the track if there is a problem with the amp.Doing the layout is fine for your own amp but someone else may like your schem but might do the layout different. The way i have done this layout i am able to follow it but i can't say that anyone else can. I'd do the board and components in black, grounds and earths in green,B+ in red, tube sockets in yellow,pots in blue and as for long wires that cross the circuit board tag each end of the wire to help with not being able to see component value.Post layout and schem together for comparison because you can not always get all the info from the layout.Hope this will help,Thanks

[DummyLoad]

if R3 or R4 or R9 are going to dissipate any significant heat, they may melt plastic/rubber of can cap. be sure to mount them so that they have air space all the way around the part. measure the parts you're going to use and be sure the turret spacing is adequate.

have fun! :-)

--DL

[HotBluePlates]

Will this be installed in a head or in a combo cabinet?

[spacelabstudio]

Will this be installed in a head or in a combo cabinet?

I don't know.  Probably a head, but the chassis can be installed in any kind of cabinet.  Does it matter?

chris

[HotBluePlates]

I'm trying to envision if the tubes will be upright on the largest surface of the chassis (head mounting or blackface style), or pointing downwards while installed on a side of the chassis (as in a tweed combo).

[spacelabstudio]

I'm trying to envision if the tubes will be upright on the largest surface of the chassis (head mounting or blackface style), or pointing downwards while installed on a side of the chassis (as in a tweed combo).

They will be upright.  I've layed it out such that if it is head mounted at the bottom of a head cabinet, with tubes and transformers pointing skyward, the controls will be in standard Fender order starting with the input jack and volume on the left on to power switch on the right.  So the layout drawing is looking down on the chassis when it is flipped upside down like it would be to work on it.  If I later decide to do a combo mount, the controls will be backwards from what I'm used to, but I suspect I'd get over it.

Chris

[HotBluePlates]

Gotcha.

The components for V1 should be shifted right (as looking at your layout) and/or the socket shifted left. you should be able to line up the components essentially with the section of the tube they go to. Look at a blackface Fender input tube for an example. The concern is having the wiring cross over itself needlessly.

For R22 (not a bad idea, by the way), you'll need to make a good mechanical connection to the wire before soldering. It would be nice to have a mechanical standoff to support the free end of that resistor, or a solder lug on the board to support that end of the resistor. If you do place a lug or eyelet on the board, you'd like the board to be fairly close to the socket, so that the resistor body is as close as possible to the grid pin itself.

In your text notes, you remark that C11 may not be used. Yes, you don't normally see it in amps where trem is injected at this point, because the cap will tend to bypass the trem signal to ground, nullifying the effect. But wait! You can still use a bypass cap, it will need to be much smaller than typical. Use a 1-2uF bypass cap. The trem signal is going to be about 10Hz or less. The resistance the cathode bypass will work against will be somewhat less than 1.5k, because the cap also sees the internal resistance of the tube and the plate load (in series) in parallel, with the cathode resistor. Call it ~1.4k. Against 1.4k, a 2uF cap is 3dB down at about 113Hz, which should keep it from rolling off the trem signal.

For the feedback wire from the speaker jack, I'd run that wire along the rear apron, tucked into the corner formed by the top and back of the chassis, then turn 90 degrees to the right to come right up to R18. For a starting point for the feedback resistor values, steal the AA764 Champ values of 2.7k for R18 and 47 ohms for R19. Tinker R18 up and down to find the setting you like. Using 47 ohms for R19 keeps its effect on tube bias insignificant.

Your component spacing for the preamp section is probably unrealistically tight, unless you're going to be using small metallized polyester caps, such as the Xicons Doug sells. If you have some parts on hand, you may want to measure and/or physically lay them out, and see how your spacing really shakes out.

Other than that, you can probably get some space savings if you rearrange the trem circuit. Look at blackface Fender layouts for some ideas, but recognize they used ceramic disc caps which allowed a tighter spacing than polyester or polypropylene caps.

[spacelabstudio]

Gotcha.

The components for V1 should be shifted right (as looking at your layout) and/or the socket shifted left. you should be able to line up the components essentially with the section of the tube they go to. Look at a blackface Fender input tube for an example. The concern is having the wiring cross over itself needlessly.

For R22 (not a bad idea, by the way), you'll need to make a good mechanical connection to the wire before soldering. It would be nice to have a mechanical standoff to support the free end of that resistor, or a solder lug on the board to support that end of the resistor. If you do place a lug or eyelet on the board, you'd like the board to be fairly close to the socket, so that the resistor body is as close as possible to the grid pin itself.

In your text notes, you remark that C11 may not be used. Yes, you don't normally see it in amps where trem is injected at this point, because the cap will tend to bypass the trem signal to ground, nullifying the effect. But wait! You can still use a bypass cap, it will need to be much smaller than typical. Use a 1-2uF bypass cap. The trem signal is going to be about 10Hz or less. The resistance the cathode bypass will work against will be somewhat less than 1.5k, because the cap also sees the internal resistance of the tube and the plate load (in series) in parallel, with the cathode resistor. Call it ~1.4k. Against 1.4k, a 2uF cap is 3dB down at about 113Hz, which should keep it from rolling off the trem signal.

For the feedback wire from the speaker jack, I'd run that wire along the rear apron, tucked into the corner formed by the top and back of the chassis, then turn 90 degrees to the right to come right up to R18. For a starting point for the feedback resistor values, steal the AA764 Champ values of 2.7k for R18 and 47 ohms for R19. Tinker R18 up and down to find the setting you like. Using 47 ohms for R19 keeps its effect on tube bias insignificant.

Your component spacing for the preamp section is probably unrealistically tight, unless you're going to be using small metallized polyester caps, such as the Xicons Doug sells. If you have some parts on hand, you may want to measure and/or physically lay them out, and see how your spacing really shakes out.

Other than that, you can probably get some space savings if you rearrange the trem circuit. Look at blackface Fender layouts for some ideas, but recognize they used ceramic disc caps which allowed a tighter spacing than polyester or polypropylene caps.

All good feedback!  Thanks!  As far as routing the feedback wire, I was actually planning to stuff the heater wires up in that corner.  Maybe the feedback wire could be run along the outside of the chassis?  I don't know.  That seems like the weakest link in this layout.

I'm not sure if you can get sense of the real spacing of components on the board since it's hard to know what the scale actually is.  I did measure most stuff and I *think* it will all fit.  I will try to figure out how to get a 1:1 scale printout so I can actually lay some components down over the printout to make sure they actually fit.  I'll take a look at the trem circuit layout again, too.  That was a little frustrating trying to figure how to pack all of that in in a way that was still somewhat tidy.

Any notion whether the filter cap drain/status LED is a stupid idea or not?

I'll make some revisions and repost.

Thanks!
Chris

[HotBluePlates]

As far as routing the feedback wire, I was actually planning to stuff the heater wires up in that corner.  Maybe the feedback wire could be run along the outside of the chassis?  I don't know.  That seems like the weakest link in this layout.

Well, you can do the heater wiring two ways.

In a tweed style build, where the tube sockets are on a plane 90 degrees from the board, the heater wiring is tucked in along the lip at the edge of the chassis, because that keeps that wiring as far away from the grid wiring as possible. In old Fender, the grid wires often fly through the air from the board to the grid pins, where the other wiring is dressed against the chassis (as is the heater wiring). That reduces the chance of radiating hum in the grid wires. In this type of amp, you lay in the heating wiring before installing the board and doing any wiring. This allows for easy installation of the wiring, and eventually the board and its wiring.

In a blackface amp, the sockets and the board are on the same plane. In this case, the board and associated wiring is installed first, and dressed flat against the chassis. You install the heater wiring last, twisted and up in the air, dropping down just to connect to the socket filament pins. This method keeps the heater wiring again as far as possible from the grid wiring, while keeping grid wires as short as they can be.

That's why it's important to know exactly what kind of chassis you'll be using, and how you intend to mount it. Other decisions that seem insignificant may be based off of these choices.

[HotBluePlates]

Any notion whether the filter cap drain/status LED is a stupid idea or not?

Seems like a good combination pilot light and shock hazard warning to me.

Oh... the comments I made in the thread on the "5w Amp" regarding the power supply will apply to your amp. If you have any bass response at all in your speaker, you'll want an additional stage of filtering before the output tube plate and OT.

[spacelabstudio]

So I started out to trying to avoid crossing the leads to V1 and wound up significantly redoing most of the layout.  Same urls, which I'll reproduce below:

http://spacelabstudio.com/amps/se1/se1_layout.png
http://spacelabstudio.com/amps/se1/se1.txt

I didn't realize how difficult layout was until I tried to do one.  I think this revision is significantly tidier than the first.  What do you think?  Any obvious problems?

Chris

[HotBluePlates]

You may want to add the new layout to your site again. I'm still getting the original layout, even after refreshing the page.

[spacelabstudio]

You may want to add the new layout to your site again. I'm still getting the original layout, even after refreshing the page.

Fixed.  Sorry about that.

Chris

[HotBluePlates]

Why have multiple components find their ground at the back of pots, when there is a handy ground buss right on the board? At least, I'm assuming that's what's going on with R14, R21, C5 and C9. If the pots have a ground wire along that pot backs, just connect that buss to the board ground buss at one point. Follow Doug's board making instruction in the Library; you'll want to install that pot buss with the pots not bolted to the chassis (which acts as a heatsink), but still spaced properly for neatness.

When I suggested copying Fender's trem layout, I was thinking about how C6, C7 and C8 take up a lot of space. You're planning on using disc caps, so why not arrange them in a line from the top to the bottom of the board, like a VibroChamp? Your circuit has one extra cap, C9, which is not in the VibroChamp circuit. I'm sure you can find a creative way to place it.

*If* you present drawing has the components to scale (not sure they are), it sure would be nice if C5 was physically between the trem circuit and the preamp. If you don't use a Sprague Atom, C5 will likely be smaller. Either way, the idea is it would allow neat, straight lines. It's probably a dumb idea for me to propose, but what if R11 spanned from left to right to C5, as it does now, but R12 then moved from right to left? You'd probably want to add 2 turrets in that case, to allow R12 to be positioned next to R11.

I don't know... probably a dumb idea.

Speaking of more dumb ideas... R18 just seems brain-damaged to me. I know it's on the original DeArmond trem circuit, but it seems to serve no useful purpose. If you simply copy all of Fender's VibroChamp trem circuit, you can loose at least 2 parts (C9 and R18), and copy a proven layout.

[spacelabstudio]

Why have multiple components find their ground at the back of pots, when there is a handy ground buss right on the board? At least, I'm assuming that's what's going on with R14, R21, C5 and C9. If the pots have a ground wire along that pot backs, just connect that buss to the board ground buss at one point. Follow Doug's board making instruction in the Library; you'll want to install that pot buss with the pots not bolted to the chassis (which acts as a heatsink), but still spaced properly for neatness.

You might be looking at it funny.  The one and only ground buss is on the board.  Of the components you mention, R14 does not go to ground and the others all grounded on the buss.  Where a pot needs a ground connection I'm running a wire to a point on the ground buss.  I'm trying to more or less follow Merlin Blencowe's grounding scheme, except I'm just using a buss for convenience and not going overboard trying to do a bunch of local star grounds.  This is why C5 is positioned where is too, since the trem and preamp circuits are both downstream of that filter cap.  The filter caps are drawn larger than they will be in real life.  I hadn't chosen specific parts, yet, so I tried to design for the worst case scenario.  I was looking at F&G caps the other day and while they are significantly shorter than the drawing they aren't that much narrower.  For other parts I tended to round up to the nearest even number of millimeters for ease of using the grid, so most things are at least subtly oversized, in part to force me to leave enough room.

I'll take another look at the trem layout.  If there's no obvious problem with it, though, I might call it good enough to prototype.  As far as the extra parts in the Dearmond circuit, since I'm pretty much just cargo culting that circuit anyway, I might as well leave them in for now.  It will be a lot easier to experiment with removing them from the board than adding them to a board not designed for them, assuming I want to see if there's any practical difference in the effect.  I do have a Vibrochamp, so I'm familiar with it's tremolo sounds.  Will the Dearmond circuit sound different?  Probably not, but maybe?  I'll found out I guess.

Thanks for looking it over.  I appreciate your help.

Chris

[HotBluePlates]

Why have multiple components find their ground at the back of pots, when there is a handy ground buss right on the board? At least, I'm assuming that's what's going on with R14, R21, C5 and C9. If the pots have a ground wire along that pot backs, just connect that buss to the board ground buss at one point. Follow Doug's board making instruction in the Library; you'll want to install that pot buss with the pots not bolted to the chassis (which acts as a heatsink), but still spaced properly for neatness.

You might be looking at it funny.  The one and only ground buss is on the board.  Of the components you mention, R14 does not go to ground and the others all grounded on the buss. 

Then your Depth pot has 4 lugs? R14 has a wire running to either a 4th lug, or what I assumed to be the case of the pot.

The trem can work as-is. I haven't heard the DeArmond, so I don't know if it will sound different. I suspect that it won't.

[spacelabstudio]

Why have multiple components find their ground at the back of pots, when there is a handy ground buss right on the board? At least, I'm assuming that's what's going on with R14, R21, C5 and C9. If the pots have a ground wire along that pot backs, just connect that buss to the board ground buss at one point. Follow Doug's board making instruction in the Library; you'll want to install that pot buss with the pots not bolted to the chassis (which acts as a heatsink), but still spaced properly for neatness.

You might be looking at it funny.  The one and only ground buss is on the board.  Of the components you mention, R14 does not go to ground and the others all grounded on the buss. 

Then your Depth pot has 4 lugs? R14 has a wire running to either a 4th lug, or what I assumed to be the case of the pot.

The trem can work as-is. I haven't heard the DeArmond, so I don't know if it will sound different. I suspect that it won't.

I should mark that better.  It's a push/pull with an SPST switch.  The two lugs that are standing out are the SPST switch and the three on the body are the normal pot lugs.  Of which only two have wires attached.

Chris