LM386 Instability Issue

[audioguru]

Why don't you ues an ICL8038 or an XR2206?

I used the Intersil ICL8038 about 37 years ago. It is obsolete now and has not been made for many years.
I have never seen a similar Exar XR2206 that was available up to about 10 years ago.

Of course some people hoard these parts and sell them later on e-bay and get rich. I should have done DAT.

 

[chemelec]

I used the Intersil ICL8038 about 37 years ago. It is obsolete now and has not been made for many years.
I have never seen a similar Exar XR2206 that was available up to about 10 years ago.

Of course some people hoard these parts and sell them later on e-bay and get rich. I should have done DAT.

Well if he wanted one, I could send him an XR2206 for just the Shipping Cost.

As to there Cost, They Never were CHEAP, (They were $6.00 Wholesale in 1988) but that is not an issue for me as in a few years, All my parts will be going to the Garbage Dump. I have no Choice, but WHAT A WASTE!

 

[audioguru]

My son met a guy who used to work as a technician at a store chain that sells home audio equipment. This guy was moving and needed to get rid of hundreds of "unrepairable" products returned by customers. My son took them all for free.
I saw many working MP3 players, digital cameras and stereos that were returned because the customer wanted a different COLOR or something. I took 4 defective MP3 players and made one good one from their parts.

Alas, I also threw away many of the products but I kept and repaired a huge, heavy 5 channels power amplifier that has a crushed corner (I guess it was dropped). Most of the stereos were Cheep Junc and I am glad I threw them away.

 

[chemelec]

Hi audioguru, These are All NEW Parts.

My Stock Exceeds $100,000.
Unfortunately I live in a Very Small town (Pop 335) and shipping LARGE Quantities of this stuff to outside of Canada is becoming more and more of a problem. Mainly due to CUSTOM PROBLEMS.

Even within Canada is getting difficult because of Higher Shipping costs now.

 

[breakshift]

They both appear to be discontinued. I've looked for chips on RS, Farnell, Mouser etc that would do this for me but I haven't found any - I'm most likely looking in the wrong places. Are there modern equivalents to those two chips?

Would it be possible to use the oscillator in the NE612 in a Colpitts configuration, and achieve the frequency range by using a varicap?

 

[chemelec]

Yes All Produced Sine, Square and Triangle Waveforms, Up to 1 Mhz.
They are All Discontinued now.

"Mouser" has three of the XR2206's "Function Generators" in stock and they want $75.00 EACH.

My Offer was FREE, Excluding Actual Shipping Cost.

The ICL8038 and the MAX038 were also function Generators.
But I don't see Any of these Available Anywhere.

I am Not aware of any other options.

 

[rogs]

They both appear to be discontinued. I've looked for chips on RS, Farnell, Mouser etc that would do this for me but I haven't found any - I'm most likely looking in the wrong places. Are there modern equivalents to those two chips?

Would it be possible to use the oscillator in the NE612 in a Colpitts configuration, and achieve the frequency range by using a varicap?

I can't help feeling this mixer is getting to be way too sophisticated for such a simple task!...

I built a heterodyne bat detector about 10 years ago, and just used an 'off the shelf ' bog standard 4052 CMOS multiplexer as a switching mixer. Because the LO for a switching mixer is a square wave, only the 'signal' input - and its inverse - need to be linear. Much simpler !
(For the maths, see page 6 of this paper:
https://www.analog.com/library/analogdialogue/archives/43-09/edch 4 rf if.pdf )

The rest of the circuit (including the oscillator) used TLO72 opamps. Using an oscillator in that configuration removes the need for power supply regulation to maintain oscillator stability.

I've attached a handwritten sketch of the mixer ..(in a previous thread I spent ages drawing out a circuit for someone with one of these computer 'schematic drawing packages', and no one even bothered to read it... so a pencil sketch it is for the present!)....

It's just the mixer and oscillator part of the original circuit. The preamp is probably not really what you'd be needing with your MEMS mic... 10 years ago I was restricted to using an electret mic - with a less than ideal ultrasonic response - so the preamp was built to compensate for that.

In the UK (and as you mentioned receiving BBC I'm guessing you're in the UK?..) the main frequency band of interest is likely to be from 25KHz (noctules) through 50 - 55Khz (pippistrelles). Serotines and Daubentons are also covered in that range.
The Horseshoe bats are up around 120 KHz, but they are so rare I didn't bother to try and cover that range, with a less than ideal microphone!

There's a small mp3 here : www.jp137.com/las/soppips.mp3 that I made a couple of years ago of some local soprano pippistrelles... the oscillator for this recording was set around 55KHz.

Another thing I did differently (although I did fit an LM386 at first) was to ditch the power amp. You can drive headphones directly from the output of a TLO72, when the ouptut is fed via a low(ish) value output resistor ... say 100R. (any less and the amp might oscillate, with the reactive load presented by the phones.

In practice, I fed the output from the detector directly into a cheap audio recorder (I use a Sony like this model:
https://www.amazon.co.uk/dp/B00BOK8ZB6/ref=dra_a_cs_hp_hn_it_P1700_1000?tag=dradisplay0bb-21

and plugged the headphones into that. Increases the life of my single 9v pp3 battery in the bat detector considerably....

Just a few suggestions you might like to consider for your project......

 

[audioguru]

The bats are not singing to you, they are just making "clicks". So high fidelity is not needed.

Did you see the TV show about a blind guy who makes clicks and "hears" where objects are around him? The objects reflect his click sounds and his ears and brain make an image of the surroundings. Like bats?

 

[rogs]

The bats are not singing to you, they are just making "clicks". So high fidelity is not needed.

Well... that's probably true for heterodyne detectors, definitely true for frequency division detectors, and probably not true for time expansion detectors!

My first detector was frequency division, and used a simple 40KHz ultrasonic piezo as the detector. As you say, it just produced clicks.

My heterodyne detector does a little more..... some of the 'wet slap' type sounds suggest the bats are generating frequency modulated echolocation sounds, and not just bursts of random noise. The heterodyne detector records the frequency shift as it passes through the mixers audible range. It also allows for control of the detected frequency range, to be able to help distinguish between different species of bat. It is also able to detect amplitude variations. So a reasonable amount of fidelity is required.

The latest time expansion detectors record the whole ultrasonic spectrum of bat calls, with pretty good fidelity. These are then slowed down for analysis. These recordings are quite wide ranging, both frequency and amplitude wise, and so require quite high fidelity.

Research does suggest the use that bats make of ultrasonic echolocation is really quite sophisticated.
I mean, when they generate a burst of ultrasonic 'sound' they have to be able to detect the return from their own source-- sometimes in among a whole flock of bats all feeding at the same time, and each generating their own ultrasonic 'bursts'at the same time.

It's not like light, where we all just see reflections of the same source. For ultrasonic echolocation each creature generates its own 'source', and has to detect its own specific reflected response.

They're pretty smart little critters... I suspect its more than just 'clicks'

 

[jjw]

Some species "sing", they use frequency sweeps in their sounds.
There is Batlib app for Android in Google Play.
Listen sounds of Common Noctula.

 

[audioguru]

Please post an example of bat sounds that are more than simple clicks.

 

[rogs]

Have a listen to the samples here: **broken link removed**

Once slowed down, they're more like bird song, as you can hear.

There are some excellent notes here on British bats: **broken link removed**

Probably not a lot of use in Canada though!

Unfortunately, I haven't got the funds to buy a time expansion detector.... and making one would be quite a project!

 

[jjw]

This is from Youtube:

 

[audioguru]

I didn't know that bats can clap their hands! And stamp (stomp?) their feet!
I heard squeaks and clicks most of the time on those recordings.

 

[rogs]

I didn't know that bats can clap their hands! And stamp (stomp?) their feet!
I heard squeaks and clicks most of the time on those recordings.

As I say, I think they're amazing little critters, and we have probably only just begun to understand just how clever their echolocation techniques are?.......

....Seems like you're happy to stick with being known as 'audioguru,' and are not really interested at all in becoming 'ultrasonic guru' ...

 

[audioguru]

I heard ultrasonic burglar alarm motion sensors when I was a teenager and in my twenties. They were extremely loud. Now I am an old geezer and I cannot hear ultrasonics anymore or maybe they are all infrared now.
I take a small Asprin every day to prevent heart attacks and strokes and they give me tinnitus which is a faint high pitched ringing in my hearing most of the time. It is at about 16kHz.
I am amazed that many people hear only telephone bandwidth, 300Hz to 3kHz or worse.

 

[large_ghostman]

When we lived at on the farm in Devon, our house was very old. (around 300-400 years i think), and we had large colonies of different bats living in the loft and some room beams, it was common to see bats in the house. At night before it got really dark, you could stand on one stair case and hear the bats clicking, just before they flew out of the gap in the roof.
Dad thinks we had 3-4 different species there. When we moved this year I was sad I wouldn't see or hear our bats again and the people who brought the farm hate bats .
But since being here a few months I have seen very large bats in the evening, swooping in our courtyard! They probably live in the out buildings, Its a old house but I am pretty sure no bats live in the actual house, but at least we still have bats around! All of our family loved the bats we had, and its great that we have some here as well! I have held a few bats that we found in the old house crawling around, they are fabulous creatures, maybe a bat detector should go on 2014 project list

 

[audioguru]

There are some fairly old trees around my home and a huge river not far away. We had many mosquitoes and many bats flew around in the evenings eating them.
Then the city put larvacide in all the rain drains and ponds to kill the mosquitoes and now them and the bats are gone. The larvacide does not affect birds and animals.

 

[unclejed613]

first of all, if you are feeding the internal oscillator of the NE612 with a square wave external oscillator, that's not the way it's supposed to work. you can vastly simplify the whole thing by eliminating the external oscillator, and building a tank circuit for the 40khz signal to be generated internally in the NE612, or use a 40khz crystal (much more stable) here's the data sheet: **broken link removed**

 

[rogs]

you can vastly simplify the whole thing by eliminating the external oscillator, and building a tank circuit for the 40khz signal to be generated internally in the NE612, or use a 40khz crystal (much more stable)

A fixed frequency local oscillator is not really suitable for this project, as it needs to be adjustable over quite a range, to be able to 'detect' the different species of bat.
Hence my simpler circuit suggestion, in post #27, for a switching mixer, which does use a square wave as the local oscillator.
Although the frequencies involved here are quite high, from an audio point of view, they are extremely slow from a radio point of view (with the possible exception of the US Navy's ULF radio experiments ...)

A simple switching mixer, with a variable local oscillator made using CMOS logic and op amps is a much more suitable (and cheaper!) approach in this instance.

All IMHO of course... but I have built it, and it does work!