Op-amp behaviour with a failing battery

[...]

Yes, it is played with wooden hammers that look like back-scratchers.

You can also see the prototype mic in that picture

What a wonderful legacy to remember him by.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

Yep, I know.

Yes. Here is the one he made for me...

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

[...]

Can't listen to MP3 here at work but then how about generating an audible warning? Either a buzzer in the little box or some sort of "pock, pock, pock" in the audio. Musicians are usually pretty good at hearing an ever so slight distrubance while playing. IOt doesn't have to sound all the time, just long enough or at intervals for them to notice that the battery is on the last 15 minutes and they better wrap up the crescendo soon.

[...]

--
Regards, Joerg

http://www.analogconsultants.com/

A very different instrument from the hammered variety. How is it played, across the lap? Are the two strings on the left tuned in unison and played as one?

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

That might not be such a disadvantage - it's not pleasant listening.

On checking the recording of the event, I discovered that the mic had fallen drastically in gain at least an hour previous to the farting event and had been completely dead for about 20 minutes. The musicians didn't notice that!

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[...]

Not sure what kind of music, but maybe 20 years of heavy metal have made them almost deaf?

Then the solution could be: Use a voltage supervisor circuit of the homebrew kind that gradually pulls down the volume until, at around 6V or so, it's gone. No more screamy-farty sounds. If nobody notices the dead mike channel, fine. If they do then someone might just get the idea that the battery is exhausted. Or that it's time for a break and a round of McEwans Heavy :-)

--
Regards, Joerg

http://www.analogconsultants.com/

English folk dance

The mic capsules can be made to fade out at some suitable battery voltage, but that still leaves the amplifiers in circuit. I think some sort of two-transistor hysteresis circuit, which cuts the +ve rail , is what is really needed.

If the channel repeatedly cuts out and then back in again as the battery volts recover, that might be the least-bad option.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

Yes.

I have no idea. I only play the clarinet ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Reminds me of square dance which I did for a few years before I found my missus.

Just experiment what would make it go gracefully. IOW "fartless" :-)

Worst case you could use a FET that gradually shorts out the audio at the output, that is almost a guarantee for a soft quitting process. Because anything abrupt can cause a *POP* in an AC-couple circuit. For example, it literally took RF guys decades to come up with an FM squelch circuit that didn't do that.

With hysteresis you could prevent it from recovering but maybe that is useful.

--
Regards, Joerg

http://www.analogconsultants.com/

reasons

The next job I want to tackle is mic for a bass clarinet.

Step one is to go and record one with a variable-pattern mic at various positions and distances to hear which arrangement gives the most acceptable sound. Step two is to work out how to get that sound on stage without feedback.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

We do squares, longways sets, contras, circle dances - almost anything we can steal from other traditions. The music (like the dance) ranges from genteel to wild, but always tuneful and always fun.

The problem with a shorting FET is that it will either require a +ve voltage to keep it on, just when the battery is dead; or it will require a -ve voltage to keep it off, which would mean a whole lot of extra power supply components.

I've just knocked up a circuit which switches on and off at about one-minute intervals as the battery collapses and recovers. Thinking about it, I'm not entirely happy that it wouldn't cause considerable confusion if it operated during a performance.

The best results so far have come from substituting TL072s for the NE5532s, they don't have any tendency to fart and their noise level is not as bad as I thought it would be. It is almost inaudible in a quiet workshop at night and would be totally negligible compared with the ambient level onstage during a band performance. They also use slightly less battery power.

I might ask the hammered dulcimer player to give them a try and see if she notices anything amiss.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

Get TL084's... a quad... very similar to TL07x. They were my favorite OpAmp in the early '80's. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

The wildest square dances we did were in Ballard (part of Seattle). Real saloon style, beer in hand _while_ dancing. But no guns fired into the ceiling :-)

Use a very low Vth logic level FET. There are FETs that are guaranteed fully on at 1.8V and in real life their Rdson should be low enough even at 1V. That's way below the level when a low noise opamp can do a farty scream. Usually they drop silent below something like 3V total.

Or use a depletion mode device.

Usually there's so much other man-made noises at music venues that all this low-noise stuff doesn't matter. It only matters when you listen to Mozart with head phone, in your living room, in a house about 10 miles from the nearest road.

--
Regards, Joerg

http://www.analogconsultants.com/

I suspect, but can't prove, because the only models of the NE5532 available are behavioral, that it's a common-mode issue.

Thus I think Adrian's TL07x idea (or my suggestion to go TL08x) will fix the problem because both these devices have PJFET inputs.

It might be interesting to scope the "fart" to see exactly what the mode of oscillation is ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

At some voltage, the product will fail. This is why if it really really matters, you design an undervoltage lockout scheme. Fart sounds are tolerable, but what about blown out speakers?

When I was working on a contract for a land mobile company, they had a wall of blown up speaker-microphones, just to remind everyone that clicks and thumps are bad for business.

Aha ! Someone familiar with Snoose Junction !!!

boB

[...]

If I were doing it again from scratch, I would probably use a quad. The design 'just growed' from an original concept which only needed a dual. I doubt if it will ever go into even limited production, it was intended as a one-off to test a concept.

If you have ever tried to P.A. mic a hammered dulcimer, you will appreciate that :

1) No contact pickup accurately renders the pingy hammer-on-string sound which characterises the instrument. 2) The only place where a good sound balance is achieved is above the centre of the strings looking down 3) No mic can be put in the area directly above the strings because it will be hit by the hammers. 4) A cardioid mic which is high enough above the (relatively quiet) instrument to avoid the hammers will give feedback when the gain is raised.

Those were the problems I was trying to solve by generating a ribbon-like directional characteristic with a horizontal null plane to reject pickup of the loudspeakers and the wall reflections. It seemed to be very successful until the farting incident occurred.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[...]

Early in the design procedure I should have compared the noise level of the op-amp with the signal level from the capsules in the practical situation of a band performance; that would have saved me a lot of bother. For an individual mic or pickup in the average P.A. situation, anything better than 50 dB S/N ratio is overkill.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[...]

I seem to remember that it looked like a low frequency square wave with a lower-amplitude higher frequency oscillation imposed on one side.

It varies with the supply voltage and the external components. I have a pair of powered loudspeakers which use NE5532s in the crossover and they give a much more discreet fart when their power is cut:

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk