@analogueoutput
thanks for posting that it’s cool as bro, even a bit of Aussie vernacular, Bugger All, Bee’s Dick (in reference to base current). No Wucka’s* Whatsoever (in terms of subsituting BJT, Darlington pair, MOSFET or Sziklai Pair to get your desired ripple reduction).
Great ideas for the budding PSU designer!
Winner winner, chicken dinner!
*No Wucka’s is short for No Wucking Furries (swap the F and W)
I’m using Eurorack Switched-Mode Power Supply (SMPS) – V1.2 | Electronic things… and stuff
It’s a bit more expensive than a FC power supply. But it can produce 1.25 Amps per channel. And it uses a more commonly available DC laptop style brick.
Also the person selling them is a nice guy and helpfull.
Hi folks
I’ve noticed commercially available power supply modules run off 12 V DC
What method do they use to generate the negative rail ? I couldn’t find schematics online
I’m assuming they boost the voltage then divide it , but I feel like that would reduce the current available
I read about dc dc convertors but they seem to have low current output
Seperate related question - could you take a 24V laptop supply ,voltage divide and use an op amp to set up the ground rail ?
Most wall wart powered Eurorack supplies use a 12 VAC adapter. The AC is split between two rectifiers, one for the positive rail and one for the negative.
http://musicfromouterspace.com/analogsynth_new/WALLWARTSUPPLY/WALLWARTSUPPLY.php
Commercially available , as in preassembled modules
The ones I’ve seen for sale use a DC input
After much searching I’ve found this art ,which looks useful
Well, that’s embarrassing. I have that power supply, and I was thinking it was 12 VAC input.
Ha! No worries . That article I linked to is really interesting, it seems like the inverting charge pumps are a likely method
According to this article:
it appears that inductor-based switchers are preferred for loads that require more than, say, 50–100 mA
I assumed the low current requirement for charge pumps was for situations where you are boosting the voltage rather than inverting it, but having had a quick look at the LTC1144 datasheet it seems that it is just a limitation of method
I’ve gone down a real rabbit hole here! Off to read about inductors!
I found this but the IC costs about the same as the entire build cost of my half wave rectifier circuit.
Yeah - it is indeed a rabbit hole. I went down it rather hard, to the point of winding my own transformers:
It was a SG3525 PWM controller driving IRF540 MOSFETs through a transformer, rectified and filtered into LM317/LM337 linear regulators. This took 12V DC in and gave me dual rails out - around ±18V under no load, but that could be adjusted.
What did I learn about all this malarky? It’s no fun and you can get off the shelf modules that are a lot easier to work with. In the end I settled on some Recom RAC20-12DK modules - you can see it in the case here:
Sure - it hooks up directly to mains voltage, but it was a lot less faffing about.
Good luck whichever way you decide to go.
Cheers
Excellent work!
I’m currently using half wave rectification of a 12VAC supply and will continue to refine that for a while. Eventually I’d like to go to something that runs off DC. This is largely due to being captivated by a live performance I saw under some railway arches a few years ago, which got the idea of building a battery powered option in my head!
This is largely academic curiosity to be completely honest
My dreams were initially of some kind of car battery powered rave bike-trailer.
In any case, for it to be a proper academic exercise I suppose I should cite my sources:
Cheers
Thanks so much this is all good stuff x
So after looking at some power supply options, I’ve decided to get the RT-65B. The case size I’m planning is 2 rows at 80cm in length… so 16 modules more or less
My question is will I be expecting to max this out? I figured I could get a second one and wire them in parallel… But based on the power requirements (for the eurorack versions… but surely theyre the same? Lol) for the grr filter and 1222 vco, one should be enough right?
Clearly the current requirements are all over, but loosely it’s maybe about 20–30 mA on each rail per module, though a lot more on the +12 V rail for the VCO (I presume that’s due to the Nano). 16 modules at 30 mA each is about 500 mA, add some more to that on the +12 V side for each VCO. Depending on your choice of modules of course.
The RT-65B is rated, what, about 2.6 A on the +12 V side? But only 500 mA on the -12 V. So depending on exactly which modules you have, it’ll be marginal at best for -12 V.
It’s possible to set up two RS-25-12 to give you 2.1 A on each rail. If you also need +5 V you can get another Meanwell supply for that, or just add a 5 V regulator powered by the +12 V supply. Definitely more complicated than just hooking up an RT-65B, though.
I doubt you’d max out the negative rail with 16 modules. For the Eurorack versions, the power requirements are not exactly the same as the Kosmos, but not more:
#1222 is 52mA (+12V), 17mA (-12V) and #1114 is 21mA (+12V), 11mA (-12V).
@analogoutput, it’s the Nano, the display and the LEDs. They add up.
If you average all the ones in CTorp’s table except the Safety Valve and Fartbox, and then multiply by 16, you get about 300 mA. That’s maybe fine especially if that 500 mA rating is conservative or at least not optimistic. But if one of the 16 is a Safety Valve that puts you at more like 450 mA, and that to me is too close for comfort (and that’s not taking into account the comment “the safety valve (mine has bypassed 27r) pulls a lot more current than shown in table until it heats up”). So maybe, maybe not, depends on the modules. Or rather, the modules you can or can’t use depend on the PSU. I just don’t see the point of getting something that gives you a generous 2.8 A on the +12 V rail and a comparatively skimpy 500 mA on the -12 V for a synth that will draw comparable current on both. Basically most of the +12 V capacity is going to be wasted and your freedom to choose modules will be constrained by the -12 V limitation.
