I have Mutable’s Plaits and Clouds, ALM Pamela’s New Workout and Akemie’s Taiko. A few intellijel 1u modules. Doepfer VCF. Tesseracts VC Logics. Haven’t run them all at the same time on this supply yet. Akemie’s and Plaits are the ones that I notice the most problems with.
Maybe you can take a look at Modulargrid, there you can plan your complete rack including modules and get the exact consumption displayed, maybe that will help you?
Very simple. Straightforward and clean. Superb.
Oh, I’m perhaps too familiar with MG, ha.
My setup is well within the stated limits of a 7812/7912 based power supply running off of a 12vac wall wart that supplies 1000ma both from MG totals and my own measurements which is where my confusion lies. I’m experiencing something that seems wrong even though all my dc voltages are rock solid, current is within spec.
I had a long time prototyping engineer friend of mine (he designed the Atari 2600) take a look at the power supply and he thought it was quite strange that so many eurorack power supplies put such large capacitors on the output of the regulators. He said “the datasheet for 7812/7912 calls for .1uf/1uf and ceramic of film bypass caps which are great at keeping up with fast current draws. Since the regulator is actively stabilizing the power there is little need to create an extra reservoir on the output. The larger output caps present a low impedance to the regulator which cause it to work harder.”
FC microbus has 100uf on the output. Your design has 1uf plus the 47uf in parallel on the bus boards so 48uf. Most modules also have bypass caps on the power inputs which are effectively in parallel with all the other reservoirs. I’ve noticed around 10-47uf is typical on the modules themselves.
Perhaps the FC board is just pushing the regulator too hard with those 100uf caps.
I think I’ll try another version of the power supply with smaller output bypass caps to see if I get a more stable supply.
I’m building a Haraldswerk Modulation Sequencer which calls for 100uF on its rails. (The design uses a ±15V supply which feeds a pair of 12V regulators; of course I’m going to just connect to my ±12V supply directly.) Seemed odd to me but I built it that way — maybe I should change it out.
I’ve asked before if there’s really any benefit to putting bypass caps on the bus board and haven’t gotten any definitive answers. Hadn’t considered the ones on the power supply though.
What brand is this? The usual ones (TI, ST) are rated for 1.5 A (and you need decent heatsinking if you plan to pull that much from them).
“**Although no output capacitor is needed for stability,it does help transient response.(If needed,use 0.1-μF, ceramic disc).” From the TI datasheet for LM7812
Perhaps the bypass caps on the power input of the modules is needed because of the distance from the power supply in larger eurorack cases. Wonder if anyone has any theory about why all the bypass caps are used throughout?
Perhaps regulator would run a bit cooler with the .1uf caps on the output? I’m gonna try it. Maybe I’ll build two and compare the differences.
That’s for very brief peaks and low power dissipation, and not really a design parameter.
Quiescent current (i.e. no external load) is a couple of milliamps, not amps.
Yeah, you’re totally right. I gotta stop reading datasheets late at night. Sorry to post incorrect information.
What part of the datasheet would you use for design parameters? The test conditions column for both load regulation and output voltage seem useful.
Max current isn’t really a necessary figure for troubleshooting my issue since I’m having problems at a considerably lower current draw anyhow. FCs 500ma working range is a prudent current supply to work around and the limits I’ve been considering most useful.
Do you know why large cap values are used on the regulator output and again on the bus boards?
Hi everyone, was wondering if anyone could offer advice as I can’t find a post someone had with similar issues.
So after a step away from building to start making some music, I noticed some weird quirks in my machine - mostly that the oscillators (my stripboard one in particular) would play up, struggling to hold tune and sometimes gaining a ‘fizzy’ crackle, and a weird hum from the filter. These didn’t happen early on in the build, and a had a hunch it was related to power - I think I was running about 14 modules (admittedly, a lot of them diddy) off of on FC Microbus supply powered by a 1A wall wart.
I started removing modules, and lo-and-behold all the issues stopped once I took a couple out.
Now, considering I want to get these modules back in, and replace some passive modules with powered ones, I’m going to want more power. What’s the correct way of going about this? I assume it’d be wrong to power a single Microbus off of say, a 2A wall wart, and leave it there? Am I right in thinking I should build another microbus, wire it in (in series?) and then power both of those off of one 2A (or higher?) supply?
Any advice appreciated, the quicker I can resolve this the quicker I can get the grizzly new LMNC filter in my case!
Definitely add a new supply (or get a beefier one). Whether you power two from one 2A wall wart or from two separate ones, you’ll be good until the next time you need more juice.
if you connect 2 FC to one wall wart, you must connect the 2 FC in parallel like this
Great! Think I’ll just double down on the FC supply - I think I have most of the parts for it already.
Nice! This is the post I was looking for but couldn’t find - cheers Dud.
Such helpful people!
I think the fc documentation says it can be run off a 2 amp. Double check me though, caus i don’t have it in front of me
I recommend a 1000mA 12VAC wallwart for best results. This should give you ~500mA at +/-
12V and 100mA at +5V.
You certainly can run it off a 2 amp supply, but that doesn’t mean you can expect to get twice the current out of it.
Thanks, I wonder if the current would be limited in some way.
Heat. The regulators are rated for 1.5 A but they need to “burn away” the excess voltage above 12 V at the load current, and without heat sinking they heat up ~50 °C/W.
(so approximating, a 12 V AC input gives you ~16.5 V DC after rectification, so at 1 A the regulator needs to deal with 4.5 W or 225 °C heat. Without beafy heatsinks, the thermal protection turns it off long before you get there.)
Ah, that makes sense, thanks fredrik. I just dont like the idea of doing away with the barrel jack, or having multiple big chunky walwarts. Am i being dumb here, or is there a simple solution im missing here?
2 Amp is OK for a barrel jack.
5 Amp I think is too much for them, but you can buy AC/AC adaptors rated 5A with barrel jacks from Germany, so it’s probably just me… (knowing how german’s take security seriously)
If you don’t want heatsinks, put a few PSUs without them in parallel, just check you put the same phase to the same input of each. This has been discussed several times here.
soon we will be talking about water cooling and flashy led mods …