Weird audio Amp problem...

[jjj]

I built a busking Amp for the demonstration of my whistling to good music. It consists of:
1) electret Mic PreAmp,
2) ETI-486 Howl-round Stabilizer,
3) Pedal Reverb and a
4) 100W (class D) Amp running on a 11.1V LiPo.

Problem: When I turn the system on, the whistling volume is moderate, yet then when I turn it briefly off/on, the volume goes suddenly quite high and then gradually weakens.
I don't know what is causing it gradual drop? Maybe it has something to do with an old electrolytic? - Thank you in advance for any suggestions, Jo

Here's a sample of my inherited whistling, which I care to maintain (even now at age 77):
https://app.box.com/s/u68q3fn26jtwb9673xbfc751mef4wq6e

[nosaj]

I built a busking Amp for the demonstration of my whistling to good music. It consists of:
1) electret Mic PreAmp,
2) ETI-486 Howl-round Stabilizer,
3) Pedal Reverb and a
4) 100W (class D) Amp running on a 11.1V LiPo.

Problem: When I turn the system on, the whistling volume is moderate, yet then when I turn it briefly off/on, the volume goes suddenly quite high and then gradually weakens.
I don't know what is causing it gradual drop? Maybe it has something to do with an old electrolytic? - Thank you in advance for any suggestions, Jo

Here's a sample of my inherited whistling, which I care to maintain (even now at age 77):
https://app.box.com/s/u68q3fn26jtwb9673xbfc751mef4wq6e

What is the model of the 100watt class D amp and why are your turning it off/on briefly? If the amp exhibits no other problem than this I thing you could be upsetting the SMPS Oscillation pattern. If you turn it off then wait 5 or 10 seconds does it still exhibit said behavior?

nosaj

[J M Fahey]

Agree and add: divide and conquer.

Start by connecting your preamp straight to power amp, what happens now?

Post the relevant schematics, we know nothing about your system.

[jjj]

That's the Amp...

[jjj]

Ola amigo... estuve 15 anos en St'go de Chile y ahora habla un poquito chileno. (Una broma para ti: "Mi amigo que fue catholico se puso lutherano, porque le gusta el utero y ano!"

[g1]

Which LiPo are you using?
It sounds like maybe you are hearing a peak surge at turn on/off that the battery cannot sustain for long.

Or maybe you need the 'voltage up pcb' mentioned on the amp picture you posted. What is your speaker impedance?

[J M Fahey]

That amplifier can put some 15W RMS into 4 ohms, being optimistic.

That said, with good speakers itīs quite acceptable.

Measure the battery while playing, IF itīs poorly charged and drops heavily under load, that may explain your problem.

If not,there is a problem with your mic preamp, we need its schematic.

For now, forget the other pieces of equipment (anti feedback and reverb) .

Te felicito, recordás muy buen Espaņol.
Claramente por haber vivido en Sudamérica y no sólo estudiarlo en la escuela

[jjj]

That amplifier can put some 15W RMS into 4 ohms, being optimistic.

That said, with good speakers itīs quite acceptable.

Measure the battery while playing, IF itīs poorly charged and drops heavily under load, that may explain your problem.

If not,there is a problem with your mic preamp, we need its schematic.

For now, forget the other pieces of equipment (anti feedback and reverb) .

Te felicito, recordás muy buen Espaņol.
Claramente por haber vivido en Sudamérica y no sólo estudiarlo en la escuela

=====================
Soy de origen aleman. Egualmente, tu ingles es incredible... por un gaucho de la Pampa. - Una senorita ofrece a un caballero un mate y el caballero contesta: "Un mate cebados con esos lindos manos es una delicia de matear!" (Esto aprendi de mi tutor de espanol.) Otro chiste: Una (fea chichona) senora de aseo pregunta a su jefe: "Que profesor, Ud. quire casarse conmigo? El profesor contesta: "Si, porque hecho me cuenta, que me viene mas barato de casarle que pagarle por hora..." Otro mas: En un negocio visto un aviso que dice: "Si algun le habla mal de mi preguntele cuanto me debe..." Mi senora Elsita es chilena y hablo cada dia en chileno con ella. Por eso tengo pratica; 'sipo' (si pues... cachai?) - Como te parece mi silbido heredado?
Back to work:
Yes, good idea: I'm going to connect my MM to monitor the 5A2 LiPo battery while turning on the PA to see if the voltage drops after a while. I also bought some Ali-Express voltage step-up circuits and will try them in case the battery has a voltage drop. The Mic Preamp is just an ordinary electret Mic circuit. Maybe el culpable es the old electrolytic condenser ...which is almost as young as I'm?
I even added a PB switch to the Mic so, that I can "revive it" again, but that causes a nasty switching noise. If I can filter this switching noise out that'll OK too, because I can always find a dead point in the melody, where I can "revive it" the volume loss again.
Of course it would be better to find the problem and finish the madness once and for all. Maybe I have to pay for an electronics technician to locate the flaw.
Well let's see how I go.

[jjj]

Which LiPo are you using?
It sounds like maybe you are hearing a peak surge at turn on/off that the battery cannot sustain for long.

Or maybe you need the 'voltage up pcb' mentioned on the amp picture you posted. What is your speaker impedance?

Thx for the helpful reply. I just now monitored the battery on turn on etc: it showed a steady 12.56V !! The speaker is a "Pioneer" 8 Ohm speaker model.

I swapped my old two transistor electret PreAmp for the below China PreAmp circuit. It works just as good, but the main problem still exist. I am now almost sure that the culprit is that 100W Class D Amp. Tomorrow I'm going to power this Amp with a voltage step-up circuit to 14 V or so and then see if the volume remains steady.

How about this dual Mic AD633 circuit, will it effectively cut down on feedback? That would be great, for then I can eliminate the bulky ETI howlround eliminator. I'm not quite sure how it effectively works.

[J M Fahey]

Thanks
Chilean ladies are beautiful, I almost married when I was there.

Now back to the hardware problems

I'm going to connect my MM to monitor the 5A2 LiPo battery while turning on the PA to see if the voltage drops after a while.
............
I just now monitored the battery on turn on etc: it showed a steady 12.56V !!

Ok:
1) please post exact battery name/model or even better a direct link to the one you have.
I googled 5A2 LiPo battery and all I found were drone capable ones, all 3.7V (so you either are using a pack of 4 or bought some prepackaged ones) and only 850mAH which is a red flag, thatīs about 1 hour autonomy feeding a 15W amplifier.

I also bought some Ali-Express voltage step-up circuits and will try them in case the battery has a voltage drop.

They canīt do magic or pull power out of thin air; if main battery exhausts and drops, "boosted" voltage also will.
If anything, it will drop even harder.
IF you need higher voltage, add an extra 3.7V cell in series.

The Mic Preamp is just an ordinary electret Mic circuit. Maybe el culpable es the old electrolytic condenser

Please post the actual circuit.
Canīt suggest you measure a couple voltage points if I donīt know what points you have to begin with.

I even added a PB switch to the Mic so, that I can "revive it" again, but that causes a nasty switching noise.

That stinks of poor biasing or voltage leakage which mess with operating point(s).
maybe in the preamp, maybe some leak misbiases the power amp ... just guesses until we see a schematic.

You wonīt have the Class D amplifier one, but post a couple sharp and well illuminated closeups, specially showing the input and output areas.

Just guessing based on poor to zero data:

* your mic preamp may be poorly biased or leaking or missing a ground reference.
That typically "starts" ... sort of ... but slowly drifts until it cuts out ... then switching ON-OFF "resets" it.

If I can filter this switching noise out that'll OK too, because I can always find a dead point in the melody, where I can "revive it" the volume loss again.

Of course it would be better to find the problem and finish the madness once and for all.

Thatīs the point.

Maybe I have to pay for an electronics technician to locate the flaw.

Ok,the Forum takes USD, GBP, Yen, Renminbi, Euro and Chilean Pesos... which one will it be ?

I swapped my old two transistor electret PreAmp for the below China PreAmp circuit. It works just as good, but the main problem still exist. I am now almost sure that the culprit is that 100W Class D Amp.

Probably.
We NEED those PCB closeups to suggest some measuring points, also show how is speaker connected.
MAYBE some DC is leaking through some input capacitor and shifts amp bias, maybe you connected speaker wrong and amp canīt stand that for long.

I >>>GUESS<<< (lacking data ) that your amp uses a floating/balanced/bridge output and you MAY be connecting speaker wrong, grounding one of its terminals.
In that case one half amp will be shorted, maybe not exploding because of internal protection but overheating and now and then muting it.

But we need more data to proceed. Entendés huevón culiáu?

Really, "you are our eyes" .....

Tomorrow I'm going to power this Amp with a voltage step-up circuit to 14 V or so and then see if the volume remains steady.

If amp drifts or self blocks or battery dies no step up circuit will help you.

How about this dual Mic AD633 circuit, will it effectively cut down on feedback? That would be great, for then I can eliminate the bulky ETI howlround eliminator. I'm not quite sure how it effectively works.

Those are very specialized circuits, very poor audio quality , they are meant for special communications problems such as tank or fighter jet crews, noisy factories/mines, etc. , practically unusable for singing.

That said, the Grateful Dead used a similar circuit, but it was fully passive and works only with dynamic microphones: they taped a second Shure SM57 element a couple inches away from the one they were singing into (practically swallowing it) , and connected both in series but out of phase.



Ambient noise was picked up by both and almost cancelled, voice got to one very strong, and quite weaker to the other, so its cancellation was minimal.
But itīs a very tricky setup.

Of course, they NEEDED it because otherwise they "might" have had a feedback problem

[Leo_Gnardo]

That said, the Grateful Dead used a similar circuit, but it was fully passive and works only with dynamic microphones: they taped a second Shure SM57 element a couple inches away from the one they were singing into (practically swallowing it) , and connected both in series but out of phase.

Ambient noise was picked up by both and almost cancelled, voice got to one very strong, and quite weaker to the other, so its cancellation was minimal.
But itīs a very tricky setup.

Any mention of the wall of sound, plus a photo, rates a thumbs up. However there's a bit of disinformation involved. Unless it was an early, cheap experiment, dual SM57's weren't used. Matched pairs of Bruel & Kjaer condenser instrumentation mics were the norm, plus a "balance box" on the stand between them. Despite the cost & reputation of those unbelievably expensive B&K's, resulting sound quality wasn't as good as you'd imagine. Some nights yes and others, well, ....... the less said the better. What to do for your band? Best stick with an old fashioned - single - SM57 or 58. One per customer.

[jjj]

Well, it sounds like everything in my little PA is bad, but I don't think it's really ...that bad. Today I changed the Mic PreAmp for a China made one and it sounds as good as my old two transistor electret Mic circuit. Yet the main problem is still not solved. I then changed the 100W Amp for another China made Amp and that too didn't cure the main problem. - So now the problem can only be in the reverb unit or the ETI howlround reducer.
It's still all the same: after I turn on the power switch of the PA, the sound volume is moderate, yet the moment I quickly turn off and on again the switch, the PA suddenly doubles its amplification volume.
To counter this phenomenon I inserted a small speaker Pot (with a tiny heat sink on it) to curb the volume. The resistance to the speaker (including its 8 Ohm speaker) is 43 Ohm. I don't think that has something to do with the main problem or has it?
So, since I cannot find the cause of the problem I'm planning to settle for next best option: a type of push button on the electret Mic, which allows me to kick start the PA every 20 seconds or so. I got to think of some kind of circuit, which first briefly disconnects the loudspeaker and then turns off and on again the power to the PA. Maybe I can even automate it with an China off/on timer?
Otherwise the setup is truly great sounding and maybe one day I might meet an electronic expert, who admires my whistling and so, he might be willing to check it out. Until then I have to put up with what is available to my limited electronic knowledge.
It was very nice having met you wonderful people... Warm Regards, Jo from Townsville in tropical N. Australia

[J M Fahey]

Well, it sounds like everything in my little PA is bad, but I don't think it's really ...that bad.

Nobody said that, it has a problem we are trying to solve.

Yet the main problem is still not solved.

Well, it wonīt solve on its own, thatīs for sure.

I then changed the 100W Amp for another China made Amp and that too didn't cure the main problem.

Read above.

- So now the problem can only be in the reverb unit or the ETI howlround reducer.

WHAT "reverb unit or the ETI howlround reducer"
You were asked to:

For now, forget the other pieces of equipment (anti feedback and reverb) .

It's still all the same: after I turn on the power switch of the PA, the sound volume is moderate, yet the moment I quickly turn off and on again the switch, the PA suddenly doubles its amplification volume.
To counter this phenomenon I inserted a small speaker Pot (with a tiny heat sink on it) to curb the volume. The resistance to the speaker (including its 8 Ohm speaker) is 43 Ohm. I don't think that has something to do with the main problem or has it?

Who knows? You have not tested it as asked to.

So, since I cannot find the cause of the problem

You havenīt even tried

I'm planning to settle for next best option: a type of push button on the electret Mic, which allows me to kick start the PA every 20 seconds or so. I got to think of some kind of circuit, which first briefly disconnects the loudspeaker and then turns off and on again the power to the PA. Maybe I can even automate it with an China off/on timer?

Ever heard of Rube Goldberg?
He consistently designed very complex setups to solve very simple problems, such as:


That said, happy whistling

[jjj]

I built a busking Amp for the demonstration of my whistling to good music. It consists of:
1) electret Mic PreAmp,
2) ETI-486 Howl-round Stabilizer,
3) Pedal Reverb and a
4) 100W (class D) Amp running on a 11.1V LiPo.

Because I already changed 1) and 4) without success, the problem can only be either in the ETI-486 Howl-round Stabilizer or the Mooer Pedal Reverb. The thing I still need to find out if the signal from 2) output is adequate or too low? If it's too low that might explain the problem.
Yet, it still is quite weird that a kick restart suddenly overdrives the signal. - This is why I won't be able to cure the problem unless I settle for the kick start mode.

Well, the last weeks I already added a PB switch on electret Mic for it and it truly helped me to maintain good performance. The only annoying thing is the switching noise... When I whistle in public I usually meet lots of people. Chance, I might meet the right person able to solve my problem. End of problem!
My head is now spinning around ...how to get rid of that ugly switching noise?
A double PB switch, where on press one contact disconnects the speaker and the other contact disconnects and reconnects the power supply. Maybe the speaker need to be disconnected a fraction earlier? It has to be happen all in about 1 second. - Life isn't' meant to be easy!

[nosaj]

I built a busking Amp for the demonstration of my whistling to good music. It consists of:
1) electret Mic PreAmp,
2) ETI-486 Howl-round Stabilizer,
3) Pedal Reverb and a
4) 100W (class D) Amp running on a 11.1V LiPo.

Because I already changed 1) and 4) without success, the problem can only be either in the ETI-486 Howl-round Stabilizer or the Mooer Pedal Reverb. The thing I still need to find out if the signal from 2) output is adequate or too low? If it's too low that might explain the problem.
Yet, it still is quite weird that a kick restart suddenly overdrives the signal. - This is why I won't be able to cure the problem unless I settle for the kick start mode.

Well, the last weeks I already added a PB switch on electret Mic for it and it truly helped me to maintain good performance. The only annoying thing is the switching noise... When I whistle in public I usually meet lots of people. Chance, I might meet the right person able to solve my problem. End of problem!
My head is now spinning around ...how to get rid of that ugly switching noise?
A double PB switch, where on press one contact disconnects the speaker and the other contact disconnects and reconnects the power supply. Maybe the speaker need to be disconnected a fraction earlier? It has to be happen all in about 1 second. - Life isn't' meant to be easy!

So lets simplify this a little bit. Remove the pedal reverb. Then see if your problem still exists, If it does remove the ETI-486.
I really think your going to introduce more issues trying to accomadate the on/off issue.

nosaj