vocal megaphone effect

[iainpunk]

the tiny speakers i have used were green, and the guy at the flea market stall told me they were 1W 8 ohm, but the DMM says 5 ohm... i suspect part of the clipping comes from the speaker being pushed out of center because of the gorilla glue i used, it expands, thus making them bottom out quicker/easier, they sure distort at lower volume than before i glued them together.

cheers, Iain

[idy]

8 ohm speakers usually measure around 5...8 is the impedance to audio, 5 is the DC resistance.

[Rob Strand]

the tiny speakers i have used were green, and the guy at the flea market stall told me they were 1W 8 ohm, but the DMM says 5 ohm... i suspect part of the clipping comes from the speaker being pushed out of center because of the gorilla glue i used, it expands, thus making them bottom out quicker/easier, they sure distort at lower volume than before i glued them together.

It's totally normal for 8 ohms speakers to measure low *DC* resistance.   The impedance at higher frequencies is higher so on average it's more like 8 ohms.   There's standards which relate the minimum *impedance* to the allowed impedance label.   It's something like the speaker impedance shall not drop below 80% of the rated impedance *within the rated frequency range*.    You don't need to include DC unless DC is within the rate frequency ranges - a bit of a cop-out but fairly practical.

For an 8 ohm speaker 80% is 6.4 ohms.

[iainpunk]

8 ohm speakers usually measure around 5...8 is the impedance to audio, 5 is the DC resistance.

oh waauw, i feel like an idiot not thinking of that.

thnx, Iain

[willienillie]

For an 8 ohm speaker 80% is 6.4 ohms.

The Fender Champ (and many others) even specified a "3.2 ohm" speaker when they really meant 4 ohms.  Sometimes older schematics will go further and list the DC resistance of the OT primary, to aid with troubleshooting I assume.

[Rob Strand]

The Fender Champ (and many others) even specified a "3.2 ohm" speaker when they really meant 4 ohms.  Sometimes older schematics will go further and list the DC resistance of the OT primary, to aid with troubleshooting I assume.

Yeah, in the bad old days there was some ambiguity about giving a single number to something that clearly varies.   For HiFi speakers the crossover can make the impedance drop even more.   IIRC, that's when the standards kicked in.

[idy]

So the two pole 1k I still like. If I increase the gain it gets too...peaky, prone to feedback. Adding a passive high pass at the end (maybe 3 pole now?) continues to help. I hear that 3 or 4 poles is not uncommon for this kind of thing?

I tried some different op amp clipping diode pedals with this, and the additional gain made more feedback again. Hard to get gravel without getting howl. Maybe layering filters before and after this kind of thing?

My headset mic is pretty generic. I think about a better (or worse?) model or maybe a physical mouthpiece....

[anotherjim]

I won't use Gorilla glue for anything because it does expand (that brands Cyano is pretty good though).
For attaching small speakers, good old contact adhesive is best for me. Evostick is what I have. I'm pretty sure that's the kind they use in cheap bookshelf speakers. You can pry it off if you have to.

[idy]

So I boxed up the 2pole 1k with a simple passive high pass tone pot after, and am getting a feel for that. The knob with max cut is around 1k also.

I am interested in building a 4pole highness, more synthy filter, sweepable freq. I found Electric Druid's page using the CEM3320...these look do-able. Druid notes that, as my searching seemed to tell, the 4pole lowpass is a "fairly rare beast."

The new part is the Alfa LV 3320, same same?
Would have to charge pump to put in a stomp box...?

[Rob Strand]

I am interested in building a 4pole highness, more synthy filter, sweepable freq. I found Electric Druid's page using the CEM3320...these look do-able. Druid notes that, as my searching seemed to tell, the 4pole lowpass is a "fairly rare beast."

Another way to do it is to switch-in a second fixed 2nd-order high-pass, then use the existing 2nd-order variable cut-off on top of that.  Try 500Hz, 700Hz, 1kHz for the fixed filter.   You will soon find out where you like the sound.

[idy]

That sounds like a good experiment, surely simpler. I guess I may end with several of these things... I could tell there was more possibility with a gain pot, I would want to choose limits for it.

Just a shout out for the Druid with this page:

https://electricdruid.net/cem3320-filter-designs/

He makes it look "easy." I remember what the old modular synth filters sound like and have always wanted those sounds. It looks like I could strip board these 3320 filters and the new chips are like..$6....

[Rob Strand]

He makes it look "easy." I remember what the old modular synth filters sound like and have always wanted those sounds. It looks like I could strip board these 3320 filters and the new chips are like..$6....

Those types of chips makes things easier.    $6 seems quite reasonable.   A single pot to control the frequency is nice too.

In the bad old days I used switch cap filters like MF10 and LMF100.   IIRC they don't need external components but you need an external clock - which *is* external components.  [Maybe it wasn't those chips that didn't need the external components].
The 3320's are likely to be better for audio.

[garcho]

Those new Alfa 3320 (and the other CEM clone chips) sound great. I'm not messing with LM13700s anymore.

[ElectricDruid]

The 3320 gives you voltage control over four stages, so you could do a 2-pole highpass and a 2-pole lowpass and they'd track up and down together. You could get the required spacing between them by giving them different capacitor values. I've no idea what would happen to the resonance in such a situation, but you probably wouldn't need it anyway.

Alternatively, the AS3350 chip is a dual SVF filter, so you could have highpsss and lowpass with separate resonance and cutoff controls for each one.

Datasheet:
https://electricdruid.net/datasheets/AS3350.pdf

Ancient application notes on the original CEM:
https://www.electricdruid.net/datasheets/EMEngCEM3350.pdf

Supply is supposed to be down to +/-3V, so should be fine on 9V.


edit: damn speelcheck agan!

[ThermionicScott]

So the two pole 1k I still like. If I increase the gain it gets too...peaky, prone to feedback.

I'd think a little bit of feedback might be acceptable for this effect.  It's a trope that anytime someone in a movie or TV show clicks on a megaphone or approaches a microphone to address a crowd, that there's a squeal of feedback first.  (Even if said devices are better behaved in reality.) 

[idy]

But feedback is a trope I don't wish to embody, if I ever get out to gig again... But I will want to fine tune the gain if I put two filters together...

Thanks for the info on voltage.  Since I want to cut I may not need all the headroom. I can see how resonating and peaking you could run out of room fast.
All I want now is four pole high pass, but I will order a handful of chips and maybe "collect the whole set." So many fancy possibilities. 21st century Meatball...

With this and my simple acoustic pre amp I have seen how I don't need lots of bands and parametric and notch and etc etc, a simple high pass eliminates most of the evil from an acoustic instrument or a voice. I remember seeing a cheesy 70's (?) acoustic preamp here that just had two knobs, gain and bass. Oh the glory of just two knobs. Give me two good knobs...

[PRR]

...I'd think a little bit of feedback might be acceptable for this effect.  It's a trope that anytime someone in a movie or TV show clicks on a megaphone or approaches a microphone to address a crowd, that there's a squeal of feedback first. .....

Not that kind of feedback. Acoustic feedback takes a part-Second to build (despite the shockingly bad trope). Totally electronic feedback builds-up NOW (or near-enough for audio band). The filter feedback is to sharpen even peak-up the filter corners.

[PRR]

DeFord Bailey played harmonica through a megaphone and, apparently, a gourd. (In days before personal microphones.)
http://theweeklychallenger.com/deford-bailey/
https://narratively.com/the-unsung-black-musician-who-changed-country-music/

-

[idy]

So I ordered a few 3320s. I am thinking it would be nice to have some 4 pole/configurable filters in stomp boxes.

I have a first try DYLC for eyes.

I couldn't find a 18pin DIP so I kludged two together. I am guessing I can edit the component now that I have started to get the hang.

I plan to first try this with 9v and see if I need more.
The CVs would be the wipers of (10k ?) pots connected 9v to ground.
I left space for an opamp if I need to use buffered VB.

Here is the Druid's rendering of the application note circuit:

[idy]

I assume I will have to experiment with slopes, stopper resistors to make the range of Resonance and Freq friendly.