Gate/Trigger Output Buffering for MCU-Based Euro Modules

How necessary is it to use buffering for gate/trigger outputs for Euro MCU-based Euro modules?

I’m thinking about starting to develop my own, starting with a basic trigger-output modules (so I’m afraid I’m going to have quite a few stupid questions).

I’ve been casting my very much untutored eye over the circuit-diagrams of some existing modules that offer ‘digital’ outputs driven by a PIC or similar chip. Of the three I’ve looked at so far, only one (the MI CVPal) doesn’t use some kind of buffer IC to protect the microcontroller. I realise the CVPal was designed to be as simple as possible, but do the 220R resistors used provide all the protection that’s necessary to prevent accidental mis-patching taking out the controller?

a|x

Personally i use a TL07x as a simple unity Gain Buffer (or if i have a MCP680x cell available i use this), protected by a 68R when using it with a PICaxe, works well because of the Internal Protection Circuitry of the PIC that prevent oscillation of the OpAmp without further Parts.

And before you ask for Input buffering i use (and i think i spotted some suspicious schematics at some MI Modules, too) A MCP 680x Rail to Rail Op Amp between 0/+5V used as unity Gain Buffer, Input protected by a 100K, pulled to GND with 1M - this configuration will happily accept anything you throw at it and only output 0 to 5V whatever you plug in.

I highly suggest not to rely on th CVpal 220R trick, this only works in conjunction with the ATMegas internal Protective circuitry and might or might not work an another Chip. For sure you will fry a PIC 18F25K22 with the CVpal Configuration.

@fcd72 Thanks for the tips. I’m going to use an ATMega328, so maybe that has similar internal protection to the ATiny84 the CVPal uses. Having said that, the ArdcoreDIY PCB uses a Arduino Nano board, and the designer added an SN74ALS541N buffer IC.

a|x

This particular module will only have MIDI In (possibly on a MiniDIN6 connector), so the standard opto-isolator should do the job there.

a|x

If you only want trigger output with 5V you can use almost any chip you are keen to sacrifice - i suggest you use a TTL 7400 for optimum style configured as just 4 NOTs - add a 50R for protection and you have a Puzzle almost no one will solve :wink: Or vintage style just a transistor.

@fcd72 not quite what I had in mind, but thanks for the tips…

a|x

Sorry, I know it’s a bit cheeky to request info about designing my own Euro modules here, but can anyone confirm whether or not it’s advisable to use some kind of output buffer or other protective measure, specifically if using an ATMega328 for a trigger output Euro module?

a|x

Its always advisable to buffer outputs from a Microprocessor because its so much more expensive than a simple TL07x. If things go wrong (and they will) you only kill the cheaper chip.

@fcd72 thanks! Do you think this would do the job, for buffering the trigger outs?

I’m planning a 4-trigger + 4 DAC output module, similar to a dual CVPal, but with an emphasis on trigger/pulse/clock outputs, rather than CV/gates (don’t wanna have to bother with CV-scaling for perfect V/Oct tracking).

I’ll be using a quad op-amp IC (would this do?) to boost the DAC outputs to the 0>5V range, rather than the -5 to +5V range I was initially planning, for the moment, at least. Do you think I’d need any additional buffering after the op-amp, on the DAC outs?

a|x

This is for driving Logic Lines and it looks it will be destroyed if someone plugs the output of an beefy Oscillator into it.

Use a TL07x with the Cells configured as Buffer like this:

and add a 68R on the output for protection. This will be bulletproof.

Ah, that’s good to know. I recognise that configuration from the Barton MIDI To Gate module, now you mention it!

How about the DAC outs? Should I use an 8-cell op-amp, and use a second cell as a buffer, as above, for each of the 4 channels?

a|x

You don’t need to add a buffer after the first op-amp stage. If you do so, you have to be consistent and add a third buffer to protect the second buffer, then a fourth buffer and so on :slight_smile:

Ah, an infinite buffer loop…
Good point. I guess a single TL074 would do the job of protecting the DAC outs, then? They’re super-cheap, so I guess they can count as disposable.

a|x

[removed after RTFM]

a|x

You can actually use a protection resistor inside the feedback loop of an opamp to limit the current without having to deal with voltage drops on the output. Take fcd72’s drawing and add a 1k resistor right between the opamp output and the T-junction. That way, the opamp will keep the output at the right voltage and still be protected against over currents. (Writing from the phone, so no drawing - sorry)

… and you do that, you absolutely need a small cap between the op-amp output and V- pins (otherwise long lengths of cable with high capacitance will cause op-amp instability).

The TL074 already is protected against shorting internally, so adding a 68R in series on the output protects the protection and adding the 1K in the feedback loop (+ the cap to LP filter if you have long cables) add another level of protection so it might be a good idea to buffer all this, too……

I’m confused, now…

a|x

Just use the Buffer and the 68R as above. Works here with every abuse i can think off.

… and if there is already an op-amp after the DAC to scale / offset the DAC signal, then you don’t need to add another layer of protection. Just make sure that the output jack is not directly connected to the op-amp output.