[sdiy] MIDI output current and MIDI-driven gadgets

[Roman Sowa]

That 220 ohm resistor was also put there to prevent just that - 
harvesting power from MIDI OUT. Also if you accidently cut the cable or 
whatever happens to short the wires, it will not blow up the instrument 
but limit the current at roughly 20mA.

I have seen MIDI interfaces where the 5V line was literally turned off 
when there was no MIDI byte sent. So no harversting there at all.

There are manufacturers in organ world that add 2 power lines on pins 1 
and 3 of the DIN jack, and sometimes provide 7-pin DIN jack with power, 
but the power is on 1-3 anyway. Just to show the user it's not MIDI so 
hey must use their product only, while in fact it is pure MIDI with 
added power.

Roman

W dniu 2026-06-04 o 12:52, Mattias Rickardsson pisze:
> Hi folks,
> 
> Do any manufacturers intentionally design their MIDI Outs to improve 
> reliability with MIDI-driven devices?
> 
> Even though the MIDI standard only specifies communication, some gadgets 
> use it for driving electronics by the voltage and current typically 
> appearing on a MIDI Out. Like CME WIDI Master etc, adding Bluetooth MIDI 
> in a plug directly on the MIDI port, and in many other hobby projects 
> and commercial products throughout the years. 
> https://www.cme-pro.com/widi-master/ <https://www.cme-pro.com/widi-master/>
> 
> The MIDI standard allows quite wide tolerances on everything: Supply 
> voltage +5V +/-10%, resistors +/- 5%. Hence, the possible voltages and 
> currents available for a MIDI-driven device can differ quite a lot, at 
> least in theory and in older equipment. Moreover, the MIDI Out can also 
> run off a +3.3V +/-5% supply with low-valued resistors, potentially 
> allowing more current if more than MIDI's 5 mA is sucked out, but also 
> never giving anywhere near 5 V (or even 4 V) in low-current situations. 
> There seem to be observations of certain external devices not working on 
> certain MIDI gear, which is not very unexpected given the out-of-spec 
> application.
> 
> For example I'm thinking that it would be possible and fully allowed to 
> choose 1% resistors near the more current-generous end of the 5% spec 
> range, and then give slightly more current and/or higher voltage to an 
> external device, while still being compliant with the MIDI spec (and 
> hopefully well-working with all MIDI In ports). There are also other 
> potential tweaks to the MIDI Outs, optimizing its use for certain 
> "illegal" operations. The driving voltage can be tweaked to give more 
> current or more voltage at certain non-MIDI-typical usecases, and I 
> guess the MIDI Out could also be designed more as a true current source 
> than the logic buffer with output resistors shown in the spec.
> 
> Now that the MIDI spec has allowed both 3.3V and TRS since some years 
> back, it should have triggered more innovation and I figured some of you 
> might have encountered some interesting concepts and design variations. :-)
> 
> Curious regards,
> /mr
> 
> ________________________________________________________
> This is the Synth-diy mailing list
> Submit email to: Synth-diy at synth-diy.org
> View archive at: https://synth-diy.org/pipermail/synth-diy/
> Check your settings at: https://synth-diy.org/mailman/listinfo/synth-diy
> Selling or trading? Use marketplace at synth-diy.org