MIDI In with a 6N138

[Wilder Amplification]

Hey guys. I've seen 6N138s hooked up two different ways in a MIDI In circuit -

Method 1 -



Method 2 -



Is one way better/more desirable than the other?

Also in method 2, I notice that while the 1K limits the emitter base current of the first transistor, there's nothing there to limit the base current of the 2nd one. Is this acceptable due to the fact that the 2nd transistor is only sinking current during the pulse period?

I've also heard from a couple of sources that without the emitter resistor on pin 7 for the first internal transistor that the leading edge of the received byte pulse gets rounded off significantly. Anyone have any insight on this?

[Brian Swatton]

The resistor from pin 7 is not limiting the first transistor emitter current, it's in parallel with the following b-e junction, so will actually increase the current a little. I think it's purpose is to bleed of any residual charge on the 2nd base when the devices turn off. Without this, as the base voltage falls to below 0.7v and the device turns off, the base is at a high impedance to the emitter, and any remaining charge has nowhere to go. The 1k ensures the base goes right to zero in good time.

I've encountered problems with transistor circuits that omit this, it can lead to unreliable operation and noise susceptability. Anything I design always has something to address this, it doesn't need to be as low as 1k, only a tiny current is required to quickly bleed off this charge, I usually use 100k or so if there isn't already some resistance from base to emitter.

I hope that helps.

[dumbassbob]

Jon whay aren't you ever on metroamp your one of the few people who know what there talking about

[Doomera]

I was wonder if you could help me.
First let me say I am totally amateur and don't know the lingo.
So I apologize for that in advance.

I want to know how to make the simplest, cheapest, but most effective way to make a midi input circuit that i could use to control resistance or voltage(for use with, for example, a basic 555 or 358 tone generator)
Components are not an issue, if it exists and was popular, I can get it.

And I know this might be a long shot, but I want to avoid code.

Monophonic is fine, not worried about pitch bend, or velocity I can figure that out on my own.

If you can help me, please do.

Thank you.

[Mick Bailey]

This isn't as simple as it appears. Midi is a digital format and really the most effective way is to use a programmable device to generate a control voltage (CV). There are plenty of Midi to CV designs and ready-made boxes out there, but they nearly always use microprocessors to do the hard work. It is possible to avoid coding, though - to get some idea of what's involved, see Ray Wilson's program-free design over at Music From Outer Space:

http://musicfromouterspace.com/analo...0I/midi2cv.htm

You should be aware that most devices that use a control voltage to vary pitch use the 1 volt per octave (1v/o) system, and a pulse which is used to turn on the note, otherwise you'd have a constant drone of the tone generator.

[Enzo]

Agree^^^^^

MIDI signal is digital instructions for note number, note on/off, velocity, MIDI address, etc. That all has to be translated into an output for the analog world. That pretty much requires a microprocessor and some programming.

[Mick Bailey]

In my mind it would be an easier project to use a PIC - even if the OP had to make up a simple programmer, or even get someone to program the device. The simplicity of the layout means that it could easily be built on prototype board and the lower component count reduces the chances of errors.

Compare this design (which I'm not endorsing - just an example) to the MFOS version to see how much simpler it becomes when all of the necessary steps in capturing, stripping and converting the MIDI data stream are handled by a processor instead of off-the-shelf logic;

Home Page

[Enzo]

I think in my mind there is little distinction between a MPU and a PIC, just a matter of scale. either way the OP would have to write a program, regardless of who burnt it into what. I can't imagine accomplishing it with anything less than one of those.

[Mick Bailey]

I think of MPU as a general term, but usually these days it means a high-degree of integration where pretty much everything (even the supply regulation) is taken care of in one package. PIC is a manufacturer-specific device family and I still think of these in terms of simple, low-cost DIP packaged devices, although there's now much more to the product range than that.

There are quite a few projects where someone has validated a design and published the code. Then it's just a matter of getting a USB programmer and loading the code then treating the device just like any other. No different then from clipping an op-amp into a socket.

A few years ago I worked with a guy who had a reputation for adopting new hobbies without any previous warning or experience. He'd wake up and on that day decide to breed parrots, set up a marine aquarium, build RC helicopters, or whatever took his fancy. Whatever he did, he went all-in. So one day he asked me if I had any books on electronics. I gave him a few and in less than a fortnight he'd embarked on building a very adventurous PIC-based multi-band communication receiver. He bought a programmer, load of components, solder station and whole bunch of tools. He'd never soldered before and new nothing of electronics.

So, he built this whole thing on prototype board and got it working. Right from zero.

I contrast this to the people I know who've been into electronics for years, but who would not have built that project simply because it needed some chips loading up with some code grabbed from a web site.

[Enzo]

And I admit I am one of those people. To me it is a matter of where I want to put my focus. If I want a PIC application, I have to learn how to even put a program into one. I assume I have to find and build a burner. I have to learn to write a program to scan my inputs and look up instructions and execute them. I am sure I could find out, but at this point I have zero idea what language is used. I took a year of FORTRAN 50 years ago, and that is the sum total of my programming. In this case there are apparently existing programs saving me the trouble of writing one, but that will not alwys be the case. Then I get to build my thing. All those extras I am sure are great to know, and I'd only have to learn them once, but to me, it is like walking into the kitchen and needing a nice sharp chef's knife. I COULD get a hunk of steel, and a forge...and make a knife. A useful skill, but what I wanted to do was chop an onion.

I recall in another general electronics forum many years ago when PICs were new and hot, we had one guy who suggested a PIC for absolutely everything that came up. More power to him, if he can do that. Some novice wrote in wanting to know how to combine a couple inputs to output a signal to control something. A simple circuit using a couple logic gates. The PIC guy wrote in that instead of logic gates, the OP should use a PIC. I suggested that the OP didn't even know how to use a NAND gate, why would he be up to a PIC? And sure enough, a month later, the PIC guy asked some rudimentary question about a circuit using three or four logic gates. he understood programming, but not basic logic.

SOme people learn golf, and tennis, and wind surfing, and ice skating, and drink Dos Equis, and are intersting. Other guys learn golf, period, and become Tiger Woods. When I eschew computers, I could see it as a loss, or I could see it as I am concentrating on the depth and nuance of my learning over 60 years of electronics, and someone else can be good at computers.

Clearly not everyone has the same priorities.