rootz wrote: ↑Tue May 11, 2021 3:58 pm
You might be right about those caps. I'm not too familiar with them.
I suppose the insulator might be there because the can itself is also connected to the negative poles? I don't know, but why would you need an insulator otherwise?
The power supply looks okay. On the actual board are 200uF caps for the output tube plates (parallel) and one 40uF cap for the screens. I choose the nearest value based on what you can easily obtain from Tube Town or the likes. I'm not certain about the value of the 10k resistor. I think 10k is too low. It is probably more between 22k and 47k. The latter would dissipate 230mW, so it seems possible that HAD used a 1/2 Watt part here. More logical would be 22k though. I can not make any sense of the value. I see a yellow and red band (the latter being the designator for the tolerance I believe), but the next band looks more brown than orange to me. I don't know, what do you think? It's that resistor under the large ceramic cap between the preamp tube sockets.
Those pictures came up on this forum while doing a google search for more pictures about Rick Vito's #58 (now in the possession of Keith Urban IIRC). Might have been posted by Tony, can't remember.
I request you to do the same as me: sketch all possible PSU arrangements you can think of while respecting the colours of the wires. There are multiple red wires in the PSU, but they can only go so many places for example. I'm now biased that I have the solution, so I need a new set of eyes to check my work and findings.
And let's hope others chime in here if they have got any more intel, like dreric, Tony, Max and Aaron. Bright and knowledgeable minds we need.
Oh if you know a source for more pictures of OD100-WR #58, please let me know. It's that funky one with PCB's in the attached picture.
I do believe the can is connected to the negative pole, i.e. non-insulated, but even if it were, I don't see any harm in insulating the can from the chassis - as long as the negative pole has a ground point. I'm getting my information on this, from here:
https://www.studioelectronics.biz/sunsh ... list&c=123 The gut shots of the 060 show the phenolic insulation wafers. The 124 has one as well. In the 060, I can't see the connection in the photos but the front cap is right next to a ground point - equivalent to GND1 in the 124 - so I am assuming there is a bare ground wire leading from the negative pole of the can to that ground point, just as there is more visibly on the can at the back.
Considering the node capacitances, in comparison with the 124s original scheme, 20uF per node sounds about right. Current draw, voltages being comparable it is logical to assume this is ok. Also, because of the solid state rectification, it is probably better to err on the conservative side and not overtax the charging cycle of the caps with a sudden onrush of current. Though still, 36uF could be fine. Anyway, moot point, I have the 10 x 4 can cap on the way, and a hole in my bank account where common sense used to be.
I am more than happy to review your findings, more than happy to do so. Just be aware, I am learning this stuff as I go along, and also I too, would welcome the brighter and more experienced minds to this conversation. So, you have to take my theoretical understanding of this with a hefty dose of salt. However, I plan to grapple with the mathematics of this shortly and if I can make sense of it all, I would try it that way. From the designer's perspective. At the moment, I have little information on how this works and zero experience but it seems likely that we might edge towards a more likely missing-pieces-of-the-puzzle solution, by following the maths. Key to this is a better understanding of the current loads which, as you referred to earlier, differ somewhat from the published data on the tubes. This is where your experience and simulations may be instructive. Do you have figures for the current loads of each of the tubes?
And do I have the nodes labelled correctly in my most recent sketch of your schematic?
The 10k (possibly higher k) resistor, may likely be 1/2W but could the power rating and temperature coefficient of such a resistor may have more a more subtle impact on the performance? Personally, I think 1/2W sounds ok, but given the high voltages it would see, it may be wiser in a build, to go with a higher power rating, if only for a more stable output with the least drift. Again, this is only my understanding, I can't speak from any experience here. As I understand it though, MrD seemed happy with 1/2W resistors in most cases. If it's good enough for the goose... and so on.
Last night, I couldn't make much headway when the red wires started disappearing under the boards. You're definitely ahead of me on this but I mean to resume the study, (hate it when I only get half the picture!) and so, sure, I will keep looking and thinking about it. I purposefully avoided your simulation schematic, but now I plan to use it to compare and complete the picture I started yesterday. That resistor, under the ceramic cap at the back, forget it, it's a lost cause; two bands can't tell us very much if anything useful, but the maths and the simulations and finally, the build might just do the trick. The theory and the simulations will get us so far, but to paraphrase Tony, it's time to get the iron hot. The build will be the real proof of the pudding.
those pictures? Maybe I am missing something!
