Adding an XLR/Mic input to any pedal?

[tatou]

I've been thinking about how to add a mic input (alongside a regular 1/4" TS input) to DIY stompbox circuits (particularly preamps). For example:

• The demo tape fuzz is loosely based on the old TASCAM 414/424 4-track preamps; if you add a mic input you'd get something like what these guys did
• The boss fa-1 seems like a popular DIY project; if you add a mic input you'd get something closer to the jhs clover

adding a mic/line switch, db padding, and other typical mic pre goodness (phantom power? well, not if we wanna run on 9v...) would be gravy. beyond preamps, you could also use the approach to make DIY vocal effect pedals (like this).

HOWEVER, i've found little help searching this forum or the interwebz on good ways to do this. there seem to be three main options:

• use an audio input transformer on the balanced XLR input, à la FIG1 in this tapeop article
• instead of a transformer, add a famous $5 preamp to the front on the XLR side
• simply run pin 2 to signal and pins 1&3 to ground on the XLR input and call it a day

i'm unclear on the pros/cons of these different options. the transformer seems like "best practice," but as best i can tell 1:10 "step up" audio transformers (i'm a noob so i don't know exactly what i'm talking about here) run $60+, which seems steep for the lo-fi trashy 4-track preamp project. are there other reasonable and less expensive options? how does one choose the right transformer for such a project? i'm not sure what the $5 mic pre would really add, since the rest of the circuit is, after all, a preamp... and passively converting balanced to unbalanced seems like it would lose something and muck with input impedance?

so, all that said i'm hoping someone here has tried this before or has some good advice on how to accomplish this?

[Ben N]

I can't tell what the best choice is for you, but be aware that the "$5 preamp" requires a +/- 15v power supply, and that may be the most critical part of the circuit (notice that the author of the article used batteries, avoiding any sonic issue arising out of the power supply). It may make more sense for you to build the preamp as a standalone unit (especially if you need the phantom power circuitry), with XLR in and 1/4" out, perhaps including a volume control, as your pedalboard mic input box. (If for some reason you need to convert back to balanced on the way out, this would be a good place to do it, too, and call the box "I/O".)

[JerS]

In my opinion, your option 3 is a fair solution. Let's face it - if you are running a mic into a guitar style effect, the benefits of a balanced signal are voided anyway - why make life harder on yourself. Getting enough gain from the mic is the critical step for signal to noise. You can always balance the signal on the way out of the pedal if you plan to then run the effect to a console. You can do this with a reasonably inexpensive transformer from mouser - or a chip option (THAT corporation have stand alone chips to do this work). 

[anotherjim]

I think Shure makes a Karaoke dynamic mic (PGA48QTR) with an unbalanced 1/4" jack cable (capsule is normal Shure quality but they are aimed at a market that won't pay normal Shure prices). They have a mute switch which will probably be essential messing around in pedals.
I learned something new - QTR is Shure speak for Quarter inch phone jack!

[tatou]

Thanks for the feedback all!

Part of my question, of course, is to solve a problem of running mics through pedals (for which there are many simple solutions)...

But in part, I also want to learn how to approach building a XLR/Mic input directly into a pedal... for the sake of learning (and maybe also doing). It seems like several pedal companies have tried doing this, but haven't seen it as much in the DIY scene (or as much work to find/share relevant schems)....

[tubegeek]

XLR is a connector, and inputs which use it IMPLY a few features, while it does not guarantee them.

Implication #1: low impedance source and compatible input impedance
Implication #2: "mic" level source
Implication #3: low noise design with "enough" gain
Implication #4: balanced input circuit
Implication #5: possibility of phantom power available

In other words, that XLR jack is expecting a mic to be plugged into it.

For your application, you seem to have at least two choices:

A) build a wonderful mic preamp that will beautifully raise the level of your mic signal to a level more like what is expected at the input of a guitar pedal, either inboard or outboard

B) pick and choose the features that you really need, or sorta want, or know how to/can learn how to build, or can afford.

One thing to be aware of: there are "DC-DC boost converter" and "positive-to-negative converter" chips ("charge pump" is the relevant buzzword here) that will take your 9V battery voltage or pedal power supply voltage, and make it very easy to create other voltages should you need them. For example, you can take 9V in and get +/- 9V near enough, which would be pretty close to what you'd want for a +15/-15 design. You can take 9V in and generate +48V phantom power, if you need it.

Those options exist, not saying that you need to do those things for sure.

A dynamic mic, Shure SM-57 or SM-58, for example, is not going to require phantom power, and it's a really common starting point. You'd need some more gain to interface it with a guitar pedal, and a medium impedance balanced input circuit would be just fine and dandy. A very simple balanced input circuit would do this "close enough for rock n roll" for sure. It'd be a bit more expensive to do it REALLY WELL (very low noise, very well-balanced) but that doesn't really seem super-relevant here.

tl;dr:

I'd look into using a charge pump power supply and building a split-supply balanced input like the "$5" or, alternatively, something a little nicer based on one of the THAT Corp.'s purpose-designed mic input chips, there are application notes on their website that lay it all out for you pretty completely. It'd be a project that I bet you'd get some interest in here, and some help, and there may well be projects that got off the ground already in the archives here too.

Lots of mic preamp talk at the https://www.diyaudio.com/ forum. Forum member PRR has been a participant in another forum with plenty of pro sound/DIY relevance; I can't remember what it is or find it now but I saw a lot of cool stuff there when I was researching the Abbey Road Altec compressor a few years back. If I figure out where I found that stuff I'll update this post but I have to go out shortly so it won't be right away.

I'm sure PRR will chime in with much smarter answers than mine in a little bit, this is a topic he'll be interested in for sure.

[PRR]

> i'm not sure what the $5 mic pre would really add, since the rest of the circuit is, after all, a preamp...

Guitar is 20mV to 200mV.

Microphone is 2mV to 20mV.

You need some gain. But not full mike-preamp gain. But most mike preamps have poor noise figure at low gain. You may really need a $5 preamp and then a 10:1 loss-pad.

> as best i can tell 1:10 "step up" audio transformers ...run $60+

If you have to ask, you do NOT start with Jensen. They are too good for you.

The world is full of $17 mike transformers. They are not audiophile quality or professionally robust. They are often better than guitar gear.
https://www.amazon.com/Parts-Express-Female-Matching-Transformer/dp/B0002BF1KG
https://www.amazon.com/Shure-A85F-Transformer-Female-4-Inch/dp/B0006NMUHW
https://www.guitarcenter.com/Livewire/Essential-High-Impedance-Transformer-Adapter-1-4-TS-to-XLR-Female.gc
These ARE transformers. You can also buy adaptors without transformers, which will not work in all/most cases.

The transformers do not supply Phantom, once a pro technology, now super common on low-price mikes. You look for a Phantom Power adapter. Ha! They used to all be stereo, but the downsizing of "studio" has brought us one-channel PP boxes.
https://www.amazon.com/Microphone-Phantom-Power-Supplies/b?ie=UTF8&node=11974571

[tubegeek]

Paul: what's the forum I'm thinking of with the DIY studio gear projects? Some really nice ones?

[tatou]

thanks tubegeek and PRR!!

to be clear, i was never considering phantom power for these projects. just dynamic mics.

after some more research, i found this schematic for the Tascam 488 (8-track descendent of the 414/424). here the mic/line inputs appear to be unbalanced, and the first 2 channels (the other 6 are only line inputs) have an opamp gain stage trim control that, as i read it, provides 34-3,400 volts (so 30-70 dB?) of gain. the jacks aren't XLR, so i don't know if it was "assumed" you'd be using an external preamp/transformer first (like those $17 adapters?), but for a lo-fi scenario like that, i'm not sure the phase balancing matters much? i saw those $17 adapter-transformers (and probably should just get one).

but just to keep thinking/learning, are those $17 transformers available simply as stand-alone parts that can be used internally if i wanted to add an XLR input to a preamp pedal? or are the only available as adapters?

or... do the transformers need to be "1:10" (as in the DIY mic preamp schematics i've found)? or would a cheap 1:1 transformer like this followed by a sufficiently large gain stage do the trick? for example, using one of these cheap transformers on the XLR input and going more or less straight into the preamp gain stage of the 488 above (or even just the demo tape fuzz, which appears to be a simplification/variant of that circuit)?

i'm still not sure how to calculate impedance so i'm unclear on whether the transformer would be needed for this application.

[tatou]

tubegeek — is this the projects site you were thinking of? https://sound-au.com/

i just discovered it and there are several preamp projects, as well as balanced line driver, which all seem relevant. none of those use transformers.

would it be possible to run these circuits on 9v?

[tubegeek]

thanks tubegeek and PRR!!

Happy to help, assuming I actually did.

to be clear, i was never considering phantom power for these projects. just dynamic mics.

OK, so now you have avoided one complication. We shall never speak of it again.

after some more research, i found this schematic for the Tascam 488 (8-track descendent of the 414/424). here the mic/line inputs appear to be unbalanced, and the first 2 channels (the other 6 are only line inputs) have an opamp gain stage trim control that, as i read it, provides 34-3,400 volts (so 30-70 dB?) of gain.

34-3400 doesn't sound right. But 30-70 dB of gain seems like a reasonable value for a portastudio. I'll take a look later and see what you're seeing there.

the jacks aren't XLR, so i don't know if it was "assumed" you'd be using an external preamp/transformer first (like those $17 adapters?), but for a lo-fi scenario like that, i'm not sure the phase balancing matters much? i saw those $17 adapter-transformers (and probably should just get one).

The TASCAM stuff is in the "prosumer" category - IIRC, those inputs are designed to interface with either cheap high-impedance unbalanced mics or a line level input, maybe even a guitar directly. Basically anything you can throw at it, but not necessarily optimized for a balanced dynamic mic. Standard no-brainer solution would be a XLR- ¼" adapter at the end of a mic cable plugged into the TASCAM.

but just to keep thinking/learning, are those $17 transformers available simply as stand-alone parts that can be used internally if i wanted to add an XLR input to a preamp pedal? or are the only available as adapters?

"availability" - depends on whether you are looking to find stuff that exists as something you can order off the shelf or whether you are interested in breaking open (or re-purposing!) something like a cheap Shure mic mixer which has several 1:10 step-up transformers inside. When the 1:10 step-ups are built into ¼" - XLR adapter barrels they can sometimes be extracted easily, sometimes they are epoxied into the barrels. But I'm pretty sure you could find cheaper standalone parts than fancy Jensens - maybe look on Edcor's web site?

or... do the transformers need to be "1:10" (as in the DIY mic preamp schematics i've found)? or would a cheap 1:1 transformer like this followed by a sufficiently large gain stage do the trick? for example, using one of these cheap transformers on the XLR input and going more or less straight into the preamp gain stage of the 488 above (or even just the demo tape fuzz, which appears to be a simplification/variant of that circuit)?

A 1:10 transformer in general can do three things at once: provide "free" gain (x10 = +20 dB), convert balanced to unbalanced pretty well (better than most cheap non-transformer solutions) and convert impedance from low to high. These are mostly relevant for long mic cable runs - if your gizmo box is at the end of a 6' cable to the mic, balancing isn't super important, low impedance isn't super important, you can get by with a high gain no-transformer solution. But transformers are a very good not-so-cheap weapon in your arsenal.

i'm still not sure how to calculate impedance so i'm unclear on whether the transformer would be needed for this application.

A 1:1 (assuming it's got a center-tapped primary) could do balanced-to-unbalanced conversion but not gain, so you are right you'd need to have more gain behind that if it was your input transformer.
Mic inputs and pedal inputs are *usually* designed to be "bridging impedance" circuits these days, not so much "matched impedance" which was the standard in most retro gear. Bridging means that you are looking to preserve voltage level, not maximum power transfer. Bridged impedance is a load that is at least 10x as large as the source impedance of (in this case) your mic. Shure SM58 impedance is 150 ohms, so an input impedance of 1.5K or more would be enough. (Some folks prefer the sound of this mic into a higher impedance, we're getting into taste and preferences here.)

The unmentioned factor so far: (much) higher impedance inputs are susceptible to generating (a little of) their own noise. Resistors themselves actually cause noise, the larger value they are, the more this is true. So for best practices, a mic input that is going to have 30 to 50 dB of gain thrown onto it should start with an impedance that is no larger than necessary for the application. 10K is a typical value for a mic input. 1Meg is typical for a guitar input because guitar pickups are quite a bit higher impedance sources than mics.

Calculating impedance:
Transformers are ratio devices. Within their inherent limits, they will step up or down a voltage on the primary coil by their turns ratio, let's say 1:10 for an example, and the result will appear on the secondary coil. So 1V in, 10V out. (AC only, by the way, they don't do anything useful with DC.)

In impedance terms, the behavior is also a ratio but it is the turns ratio squared. So a 1:10 transformer will provide a 1:100 impedance ratio. The Shure mic with its 150 ohm source impedance will appear on the secondary as a 15K  (150 x 100) source impedance when connecting something else to the secondary coil. (For bridging, the "something else" input circuit would then need to be 150K [15K x 10] or more to preserve the voltage level.)

The most comprehensive easily available reference for this stuff is Small Signal Audio Design by Douglas Self.

https://www.amazon.com/Small-Signal-Audio-Design-Douglas/dp/0240521773

If you want a deep dive into these issues you'll get EVERYTHING you need there and lots more. But it's not necessary for your project.

tubegeek — is this the projects site you were thinking of? https://sound-au.com/

i just discovered it and there are several preamp projects, as well as balanced line driver, which all seem relevant. none of those use transformers.

would it be possible to run these circuits on 9v?

It's not the one I was thinking of but ESP is an absolute gold mine, I forgot to mention it so it's good you found it on your own.

Right up near the top (a NEW! project) you'll see an article on 12V in, +/- 12V out, so you can see Rod Elliott is wrestling with the same questions and answering some of them too. A 9V in, +/- 9V out charge pump can be made with a very few parts and would be pretty well suited (maybe overkill) for the kind of thing you're looking at. I'll take a look at the TASCAM schematic and report back later on.

[PRR]

> Tascam 488

"Mic" here is "high Z", transformer inside mike. Although many electrets and medium-Z dynamics work, as do SM58s (via simple $6 adapter) on loud sources.

The max gain is 97 not 3000.

Today you can use NJM4580.

But just get the darn $17 transformer!! It solves many cases. I always had a few in my bag.

For permanent installation, you get the XLR and transformer and some shielding for less than the price of shipping a Jensen. Unscrew the 1/4 end and wire to that.

FWIW, most of these things are 1:7 not 1:10. And FWIW a SM58 is nearer 300 Ohms than 150.

Doug Self does not cover transformers and is really a different field.

[tubegeek]

FWIW, most of these things are 1:7 not 1:10. And FWIW a SM58 is nearer 300 Ohms than 150.
Doug Self does not cover transformers and is really a different field.

OK but I got lots of the other stuff right!

I was referring to standard "mic input transformers" like the Jensen (or Altec "red" 15095) when I was talking 1:10 although I certainly didn't make that clear. I was going from memory and you know how well that works.... I honestly never knew the spec on the barrel adapters, thanks for that!

Doug Self: hi fidelity audio focus but plenty of great info on "pro" topics like mixer design and low noise inputs. I guess there are no transformers because they aren't close enough to PERFECT for ol' Doug's tastes?

OK how about Yamaha Sound Reinforcement Handbook then? Application, not design, but lots of great helpful info.

I suspect that the reason Yamaha started publishing that material as a stand-alone book was because people kept stealing the manuals from their mixers. I know I sure did.

Oh! Here is the site I was recalling, dimly:
https://groupdiy.com/

[PRR]

Somebody points out the 150 number is in the specs:
"Impedance Rated impedance is 150 Ω (310 Ω actual) for connection to microphone inputs rated low impedance"
https://pubs-api.shure.com/file/260007

Loads more than a few Kohms give "the unloaded sound". If you have a true 600 Ohm input (unlikely) the sound is different; again for 300r load, and you can try 150r load (but output will be less). I forget what changes about the sound, try and hear.

The difference is minor and probably negligible in guitar-cord work

1:7 iron is a penny cheaper than 1:10 and also gives a broader response. And in retrospect 1:10 was pretty steep even for tube preamps, although it made the spec look a bit better and was 3dB "free gain".

[Rob Strand]

Somebody points out the 150 number is in the specs

I read that as the tabulated specs are when driving a 150 ohm load (to standardize the load) and "actual" is the "thevenin equivalent" impedance.

[tubegeek]

Hmm. I wonder what the heck 150 vs. "actual" 310 is supposed to mean? I guess it means stick to 3K input Z just to be safe(r)?

If it has any importance, I found the 150 on two quick googles (Sweetwater and some random guy's blog that was the first hit) and I know I SHOULD have looked in Shure's own spec sheet but I didn't. Mea culpa!

This is why I like having you around, Paul - you keep me honest!

Q: NJM4580 is special because low noise? Is there anything else to recommend it in particular?

[ashcat_lt]

34-3400 doesn't sound right. But 30-70 dB of gain seems like a reasonable value for a portastudio. I'll take a look later and see what you're seeing there.

30db is approximately ~ 31 times gain, and 70db is 100 times that.  The "volts" label was wrong, but it basically checks out.

[Rob Strand]

Hmm. I wonder what the heck 150 vs. "actual" 310 is supposed to mean? I guess it means stick to 3K input Z just to be safe(r)?

Shure's definition(s):
https://www.shure.com/en-GB/support/find-an-answer/microphone-impedance-actual-versus-rated

So in short the "Rated Impedance" is just a broad impedance category and has no specific impact on any characteristics of the microphone or the specification values, and has no physical meaning!   Useless in the presence of the other impedance values.

[PRR]

Most mikes (most passive sources) are aimed at "un-loaded" use, tempered with practical engineering.

Historically we had a (37/)150/600 winding stepped-up to a high Z and loaded with an open grid.

On simple analysis the mike side is infinite Z and the 150/600 choice gives the step-up gain. In fact the transformer shunts the lows, and maybe the highs. The spec on the transformer implies the best source impedance, with 150 just a nice round number commonly found after 1938.

The SM58 is rather cynically wound-up higher Z to get a higher sensitivity rating. It is much too late to object to this cheat.

When you skip a transformer and tube, and resistors dominate the input, 1.5k or 2k are the usual targets, though I have seen much higher and some lower. (Phantom always puts 13k || all else.)

It's not that important. When you studio-mike a tambourine, you may move the mike 3 inches left for one change of effect, or 200r lower load for another change of effect. It is very likely none of the record buyers notice, but it keeps studio geeks excited.