ODS 124 - revisited for the 21C

[Stephen1966]

I pulled the 060 drawing from here - but it popped up again recently in talbany's post on the 013. Maybe I'm barking up the wrong tree with this one. There are other drawings of the 060 but without the reverb and photos with... probably better to go with the photos because I'm really having trouble squaring the circle with this one!

https://tubeamparchive.com/viewtopic.ph ... 93#p206693


Skyliner circuit? - yeah. Here's the guitar it's for. My pickups do all kinds of voodoo and it's got a piezo saddle as well.

The Constellation cropped.jpg

[Stephen1966]

Rootz, can I post your info on the screens supply mods that you sent over in a PM? Sounds like a solution we can share

[rootz]

I think this one is fun too: a cross between #002, #60 and #124.

Funny to see the dry mixer CF is stealing current from the triode in front of it (the CF in the reverb does that too!). Tony did some suggestions about the compression in the 002. Might this be it, simply a CF stealing some current and compressing one side of the wave form? This one only does it on large signal swings though. You'd have to engage PAB and FET to get there.

Ditched the CF protection to get a cleaner looking schematic. I'd still add it for safety though. Added a cap on the reverb driver to shave of some high end. Did the same with a grid resistor on U5. Both can be adjusted to taste to get hiss end/or high end in the reverb under control.

Anyway, let's name this one '4 triode reverb'. If you have the room for 7 preamp tubes, this one would be fun to try IMHO.

All three designs posted lose a bit of gain, 6dB, in the mixer resistors. That is the nature of the beast. Lowering the dry mixer resistor to e.g. 68k brings back some gain, but also shifts the balance between reverb and dry signal, like Marcus found out with his Wonderland/JM signature amp.

[Stephen1966]

You're on fire! The four triode reverb. I think that is almost certainly possible. A 24 inch chassis should do the job. I will stick with the three tube (6 preamp tube) designs though. It's a lot more flexible and will allow for a modular design as you suggested. The four tube reverb design could work very nicely but if it doesn't, it would have been a lot of work for the live and learn moment. I'm hedging my bets that one of these three tube designs will just sing, and if I hit a wall with one, I will have other options to play with. One of them, maybe all of them, will find that transparent sweet spot.

[rootz]

Rootz, can I post your info on the screens supply mods that you sent over in a PM? Sounds like a solution we can share

Yes, certainly!

That is a beautiful guitar Stephen. I see it has no headstock decal. Did you build that one yourself?

I hear you on the disclaimers. Would it be an idea to keep this thread to post ideas and make a new one for the actual build?

[rootz]

Do you have a link to that chassis? You bought it at tubetown.de, right? I can't find a 24" 65mm high chassis there.

[Stephen1966]

Do you have a link to that chassis? You bought it at tubetown.de, right? I can't find a 24" 65mm high chassis there.

Sure thing!

https://www.tube-town.net/ttstore/tt-ch ... -2019.html

It's actually just shy of 24" but with 19mm stock for the sides...

The Constellation, yeah! She's my baby

[Stephen1966]

I think this one is fun too: a cross between #002, #60 and #124.

Funny to see the dry mixer CF is stealing current from the triode in front of it (the CF in the reverb does that too!). Tony did some suggestions about the compression in the 002. Might this be it, simply a CF stealing some current and compressing one side of the wave form? This one only does it on large signal swings though. You'd have to engage PAB and FET to get there.

Ditched the CF protection to get a cleaner looking schematic. I'd still add it for safety though. Added a cap on the reverb driver to shave of some high end. Did the same with a grid resistor on U5. Both can be adjusted to taste to get hiss end/or high end in the reverb under control.

Anyway, let's name this one '4 triode reverb'. If you have the room for 7 preamp tubes, this one would be fun to try IMHO.

All three designs posted lose a bit of gain, 6dB, in the mixer resistors. That is the nature of the beast. Lowering the dry mixer resistor to e.g. 68k brings back some gain, but also shifts the balance between reverb and dry signal, like Marcus found out with his Wonderland/JM signature amp.

Thanks again rootz. This is gold! As I look at my circuit I'm looking at the 470k resistors at the top and tail of the circuit as an attenuation network. Marcus raised the mixer resistor from 120k in the Wonderland layout at the node of 68k/.02uF to 220k and that, he explained to me, solved the problem with gain loss. The Wonderland takes its signal from the volume pot before the second triode in the clean channel whereas we would pulling the signal off, after. So, I know there are some differences between the Wonderland and this circuit but I am curious to know what would happen to the gain if we used 220k resistors in place of all the 470k and then used a 68k mixer resistor on the first triode of the dry side. The Wonderland has a .02uF/1M/68k voltage divider feeding the dry signal into that side and we had that in the early two-tube designs as well. I have to say, I found the idea of raising the value of the mixer resistor to be counterintuitive, but this is what Marcus did, and it worked!

If we consider the voltage divider as a necessary attenuation of the signal after second gain stage in the Wonderland, then our circuit taking both the send and return from a single point in the signal chain might dispense with it altogether as we did in the three-tube design. So, to be clear, I'm not suggesting we restore that part of the design. However, by raising the value of the mixer resistor, Marcus achieved a boost in gain because he altered the balance of the signal in the attenuation network and the dry side of the mixer, with a mixer resistor of 68k compared with the wet side with 220k. Now, taking the dry mixer resistor down to 68k would make sense but it's the 470k mixer resistor on the wet side (and it's accompanying attenuation network) that seems to be part of the equation as well.

Finally, adding a .02uF cap en route to the Master, could that work to take off some of the high end?

It is easier to show you this than actually try to explain it...

RVB 5.1.pdf

Stephen

[Stephen1966]

... And perhaps remove R117 (68k) altogether. R113 (68k) could work as the grid stopper for V3b. Should we then increase the value of the coupling cap C63 from .005uF to .02uF??

[Stephen1966]

So, actually, something more like this...

I've played with the LNFB on the output triode as well (increased the resistor to 2.2M)

RVB 5.2.pdf

[Stephen1966]

ODSR power board v.4.1.pdf

Okay, it took longer than I expected but here is the power board with added nodes for the reverb. Many thanks to Martin Manning for getting it started with his layout. You can see his under the layers - the text is corrupted but you can find copies of his latest boards in the eyelet boards thread.

This should be to scale.

For anyone curious about the voltages this is expected to put out, rootz provided the simulation which this is based on.

https://ampgarage.com/forum/viewtopic.p ... 17#p427617

If you see any problems with it, please let me know... it isn't tested so buyer beware

Good health,

Stephen

[Stephen1966]

A little bit more...

Here is a sketch of the O60 layout compiled from Aaron's last 060 schematic the photos and a couple of layouts in Visio that are knocking around.
https://tubeamparchive.com/viewtopic.php?f=4&t=17935

060 layout sketch.pdf

A couple of questions remain... there appears to be an undocumented resistor in the feedback loop on V3. The colours in the photos are really bad so it's anyone's guess what it could be. My best guess - seen on the second page - is 180R. The only other clue that could be found was in Chris Barnett's notes (page1, I think).

Chris Barnett page 1 extract.jpg

However, 'D and E' seem to referring to the first tube. 'Vib' is perhaps an indication of the structure rather than actual values...

Also, in the photos, the FB resistor coming off the plate (R115) is the only one which seems to be gooped and I couldn't make it out at all... Aaron says it's a 470k but some say it's a 2.2M. When the signal is riding on this path with the reverb off it's tempting to think that 470k is an easier path but if anyone has any insight on this, I'd be grateful.

I am curious to find out what the cost in gain on the 124 circuit this is going to have. I guess it's not just about how loud it can go - 100W is enough to rattle the windows - but how compressed the signal becomes when driven through the reverb.

Honorable mentions to dreric and Dustin Beasley's Custom Fifty layout. Dustin located the cathode resistor and cap of V3 on his board while the photo shows them dangling off tag strips near the tube. My drawing isn't to scale, rather a representation of the layout inside the actual amp.

For anyone else building this amp, the sketches should give you a solid start. I have my own fish to fry with this one and studying the layout was instructive but I won't be using NOS parts so I'm looking for modern parts and I'll stitch my layout together when I know the component dimensions.

Juggling plates, that's easy

Stephen

[Stephen1966]

"Vib " V1b

[Stephen1966]

A bit of errata... If you use the 060 layout, don't forget to add the ground wires.

[rootz]

Hi Stephen, I had some busy days, but have been working on a full layout for the #60 in Visio. Much like the one by dreric in the other thread. It will fit the tube town chassis that is a bit smaller than the original Dumble one. The PSU board is smaller than original and the power transformer and power tubes are a bit closer to the preamp side.

I managed to straighten the picture that shows the whole preamp. That way I could overlay all eyelets on the picture in Visio and get a fairly accurate layout, with also more accurate spacing for the preamp tubes. By doing this it became obvious to me that the reverb transformer and a can cap are placed between V3 and V4.

Looking closer at the original amp, it also became obvious that there has probably been a lot of tweaking of the original amp. Hence resistors with long leads crossing other components, flux build up on the board at the 1.2meg resistor, a white wire going to one of the preamp tube plates while all others are blue, holes that have been filled and much more. Quite interesting to see.

Hopefully, I’ll have more time to post some work in progress this evening and read you new posts and comment on them. Stay tuned.