[Beginner]My first pedal ... not for a guitar

[DaPloup]

Hi there,

First of all, being new to the forum, thank you for your indulgence and patience but most of all, thank you for sharing your knowledge on this site ! 

I discovered this forum looking around for some info for an upcoming project of mine that involves not guitars but steelpans

You may know that steelpans are instruments from the percussions family ; they are usually played 100% acoustically.
However, some players do use pickups (like this one : https://www.ensoulmusic.com/ensoul-pan-pickup ) to open a new world of creative possibilities.

In the steelpan family, there are different instruments depending if you play in the high, mid or low register.
I personally play what is called "double seconds", a set of 2 instruments in the mid-range (alto).

This is where my problem starts : as my "instrument" is composed of 2 elements, I will need 2 pickups to get all the sounds I produce.
However, I would like send to the rest of the signal chain not 2 but 1 single audio signal which means I need to summ / mix my 2 pickups.

Second thing, my audio equipment is not really suited for "instrument" impedence/level. Ideally I would have to output a line level signal to be compatible with the rest of my equipement.

This is where I am at the present time, trying to figure out how to build such a pedal with basic/off-the-shelf elements.

I'm no engineer or electronics specialist, I have a small technical background and I'm eager to learn ; I've already soldered a couple of small projects in the past (nothing to do with music) and feel confident in my ability to do the job...only the design part of things is unclear to me.

If some of you could guide me through the next steps to identify the components I need and how they should be assembled, I would appreciate

[DaPloup]

From what I could find over the web, the summing side of my needs could be answered by an single inverting op-amp setup.
Something like this (with 2 inputs), should be OK if I understand well.


Assuming my pickups will be 100% similar, I don't think I do need input gain control for my setup, but I may be wrong.
In such case, the following setup should be the way to go.


If the basic elements are clear to me, I am not sure how to choose the right values for the resistors.
The only thing that is clear to me is that R1x values should be the same for all inputs as I want the same contribution for all inputs in the final signal.

As for the transition from instrument level/impedence, to line requirements ... I must admit it's less clear to me.
Not sure what I would need.
Reading this post, it seems that an additional non-inverting op-amp with single supply and equipped with a potentiometer in R2 position (see below), would be the way to go.


I just can't wrap my head around why I couldn't combine both op-amps into one single setup, changing the R2 with a potentiometer on this setup :

[antonis]

Hi & Welcome..

Despite the fact that there are solutions for your query, the easiest way is to implement a dual op-amp (same space & pin number as a single one) for summing and gain stages..

P.S.
Depending on your particular pickups impedance, gain stage bias configuration (R3 & R4) might not be suitable..

edit: Forget the above 'cause gain stage comes after mixing one..

P.S. (valid)
It should be better to implement an also inverting gain stage for output signal been in phase with input one..

[anotherjim]

It depends on the technology of those pickups - Ensoul info doesn't say. Perhaps you could ask them?

If they are piezo contact mics, they need a higher input impedance or else lose a lot of low-frequency response. This is better achieved with individual buffers before a mix stage. The series input resistors of a mixing opamp start to add hiss (self-noise) over 10k, although, if you use metal film resistors instead of carbon up to 100k won't be too bad. However, Piezo wants very high, around 3M but selecting a little lower is an easy way to filter out some low bass and tune to the "voice" of the instrument. Depends on the capacitance of the piezo disc.

They could be magnetic pickups which will make lower impedance easier to manage, and a direct connection to a mixing opamp possible.

[PRR]

If signals are weak, gain before mixing.

I have no clue what pickup is used on a steelpan. There's multiple types offered and different details. Would want more detail.

[DaPloup]

Thanks a lot for your answers and commetns.

P.S. (valid)
It should be better to implement an also inverting gain stage for output signal been in phase with input one..

Indeed, maintaining phase coherence would be a good thing as I will most likely have a parallel signal flow (relying on microphones) for the "dry" signals. That being said, most mixers can reverse phase so it's only "nice to have" but not 100% required in my opinion.

One thing that is not 100% clear to me is the inverted / non-inverted op-amp when it comes to the summing stage : all examples I could find describe an inverting op-amp for summing ... but is it a requirement or just habbit ?

It depends on the technology of those pickups - Ensoul info doesn't say. Perhaps you could ask them?

I have no clue what pickup is used on a steelpan.

Indeed, technical information on these pickups is rather limited.
I will try to get in touch with a technical contact, I've only been in touch with sales persons who do not seem to know all the details behind the pickups.
However, I also know these pickups are used by some players directly plugged into standard guitar pedals (example : https://www.youtube.com/watch?v=bPF88Y374d0 ). This leads me to think that the differences in signal between these pickups and a guitar must be limited ; maybe using standard "guitar compatible" values for the resistors could do the job ...

[antonis]

One thing that is not 100% clear to me is the inverted / non-inverted op-amp when it comes to the summing stage : all examples I could find describe an inverting op-amp for summing ... but is it a requirement or just habbit ?

It's a fundamental requirement..
Only inverting input (-) can be considered as "summing point", hence the inverted output..
(if you search for something like "Op-amp basics" you'll understand the reasons for this..)

[DaPloup]

It's a fundamental requirement..
Only inverting input (-) can be considered as "summing point", hence the inverted output..
(if you search for something like "Op-amp basics" you'll understand the reasons for this..)

You were quicker to write your answer than me finding this article ... 
Thanks.

[antonis]

Just take into account that all the above mentioned examples have to do with DC amplifiers and symmetrical power supply also..
(AC single supply amplifiers are a bit more complicated due to AC coupling caps and non-inverting input bias configuration..)

[DaPloup]

All right, first draft.

Tried to combine both "summing" and "pre-amp" schematics together just to better understand (and share) what I think I want to achieve at this stage.
Here's the result, but I guess it's not that simple 


From left to right :
- Not sure about the input section : i assume this really basic setup has a risk of feedback from one input to the other. Don't know which solution is the easiest to solve this ... a simple diode on each branch(nope, I'd better use my brain sometimes) ?
- I used the same labels (standing for values) for all the resistors of the summing section to achieve unity gain before the preamp stage
- I guess there can be an issue with the 9V supply between both op-amps ... or maybe not
- I inserted a variable resistor for the preamp so that I can modify gain as I desired

As for the real world values of all resistors and capacitors ... no clue 

Additional ideas / remarks :
- Depending on the input signals, it may be interesting to be able to switch off the preamp section ; if I just want to send the summed signal to standard guitar-level pedals for example. When turned on, the preamp section would produce line-level output signal to directly connect to a mixer / recorder. Don't really know how to achieve this kind of bypass ...
- anotherjim metionned buffers. Not sure how to deal with these. Buffered inputs/outputs (TS connectors, like this) ?

Just take into account that all the above mentioned examples have to do with DC amplifiers and symmetrical power supply also..
(AC single supply amplifiers are a bit more complicated due to AC coupling caps and non-inverting input bias configuration..)

And ... here comes the unkown 
Any good ressource to help get my head around this ? 

[antonis]

Let's make it more practical..

[DaPloup]

  Thanks a lot for your help.

I took a couple minutes to update the draft accordingly, just to make sure I understand what has to be connected to what.
Is this new version correct as far as 9v and ground connections are concerned on the op-amps ?


From the various drawings I could find, it looks like R2 and R3 are supposed to be of the same value, right ?

[DaPloup]

I also learnt today that the pickups are indeed piezo based.
Still no idea of the impedance, but I'm working on it

[antonis]

Your new version is only correct for two single op-amps configuration..
A dual op-amp only needs one Vcc (pin 8 ) & one Vee (pin 4) power supply connections..

[anotherjim]

I was suspecting piezo anyway. I saw the versions with different bass cut filters which are quite high for anything else. I presume the application needs to reject low frequencies for clarity? 
Most guitar gear has 1M input impedance. This is actually workable for a piezo, but only just. However, the shortcomings may actually suit the sound of steel drums.
Also, you know those pickups are used with guitar gear, adding extra weight to an argument for 1M impedance.

So, each pickup needs to first hit a preamp or buffer. Either one can be seen as an impedance converter. 1M input to as near to zero ohms on the output so it can drive lower mix pots and resistors in the order of 10k. The difference between the two is an amplifier can boost the signal voltage while a buffer passes the original level.

I could mention that a piezo contact mic can be DIY'd for under $5, but a proper one with an injection molded mounting means upfront tooling cost that isn't small.

[DaPloup]

Your new version is only correct for two single op-amps configuration..
A dual op-amp only needs one Vcc (pin 8 ) & one Vee (pin 4) power supply connections..

Now I understand why you added numbers on the op-amps 

[antonis]

Untill you decide about particular piezo pickups impedance, let's make a rough BOM..

R2=R3=10k - 100k
C4=10μF - 100μF (depending on R2/R3 value in reverse analogy..)
C3=10μF (negative plate to output)
R5=100k (say)
R4=10 (say again)
(max gain = 10)
C2=1μF (for 16Hz low cut-off frequency)
C6= 68pF - 100pF

P.S.
If you don't wish for a dead silent effect when R5 turned all the way down, add a series resistor (either between pin 7 and R5 or between R5 and pin 6) of value equal to R4 to obtain a unity gain for the lowest R5 pot setting..
(max gain X11 now..)

[DaPloup]

I was suspecting piezo anyway. I saw the versions with different bass cut filters which are quite high for anything else. I presume the application needs to reject low frequencies for clarity?

Indeed, the idea is to cut-off unnecessary low end depending on the instrument you have.
As I have access to 5 different types of them ranging from tenor to bass I will go for the "no filter" version and will use EQ later in the chain to correct that (unless I try to add a 3 band EQ to the list of requirements  )

So, each pickup needs to first hit a preamp or buffer. Either one can be seen as an impedance converter. 1M input to as near to zero ohms on the output so it can drive lower mix pots and resistors in the order of 10k. The difference between the two is an amplifier can boost the signal voltage while a buffer passes the original level.

I was thinking of using buffered jack connectors to keep it "simple" for me...

I could mention that a piezo contact mic can be DIY'd for under $5, but a proper one with an injection molded mounting means upfront tooling cost that isn't small.

Indeed, piezos are cheap compared to the complete pickups I linked earlier but from what I learnt, it's not that easy to get a correct signal from such piezos on steelpan : "The secret of its success of working is how the piezo is housed inside the magnetic casing". Sometimes you better pay for something that works and play music instead of spending hours scrathing your head for better results 

[DaPloup]

P.S.
If you don't wish for a dead silent effect when R5 turned all the way down, add a series resistor (either between pin 7 and R5 or between R5 and pin 6) of value equal to R4 to obtain a unity gain for the lowest R5 pot setting..

Ok, that seems to be a great alternative to the "bypass" idea I wrote about earlier in a very simple way : just turn the pot to minimum for a "clean" signal (direct send to guitar-level equipment) and use the pot on higher values for other uses.

(max gain X11 now..)

I think I like that

[idy]

If you design input impedance on the high side, you can always add (switch in) a parallel resistor to lower it for the instrument. My guitars use K&K that want to see less than 1M. They will sell you a cable with a 1M resistor soldered in tip to ground...or you can DIY.

I have tried this with a/b switchers and yes, it does clear up the bass, taking some of the load off of EQ. Reducing bass on the way in is good housekeeping. So much energy in those big waves can swamp the circuitry.

There are other ways to create a low-cut switch. The key is to get a basic thing that works and try out the options.