TL072 questions

[Torben]

Hi all,

I have some questions about an ultrasonic rangefinder project I'm revisiting. That's the project; what I'm really interested in here is the dual opamp input section.

Quick background: I have a working ultrasonic rangefinder on a breadboard. The controller is an AT89C4051 (8051 compatible) which, on its output, drives a 38,400Hz square wave through two opposing pins to a 40kHz ultrasonic piezo transducer. Another (supposedly matched) transducer provides the signal for the input to a dual opamp booster which uses a BC550 transistor to pull a pin on an 8051-compatible chip low so I can read it in software. I can get readings up to around 2 meters and I can hear the blips (one blip of around 16 38.4kHz 5 times a second) on my bat detector.

I am using the input section from an old EPE article: **broken link removed**
with some changes: I've skipped the pot on the BC550 base, replaced the LM358 with a TL072, modified some resistor and capacitor values, and put a diode from the BC550 base to ground to prevent breakdown as per the Art of Electronics--all these changes have made the thing work better for me and were done incrementally, and could still be wrong! ).

The circuit currently working on my breadboard is shown in the first attachment. The second attachment is for those who use LTSpice--but this also works in the real world.

My problem is only my understanding of the TL072 input section. It works fine as is. I just don't know why. I mean, I know I don't get enough current to drive the BC550 into saturation without the second opamp stage, but why? And what is the easiest/prettiest way to fix it?

My questions have to do with voltage supply and current:

1) Why is the TL072 even working at all on 5V in a single-supply setup? I thought from the datasheet that it needed at least 7V. I am running it on 5V, single-supply, so 2.5V at the positive input).
2) I am using both stages of the TL072 to reamp the input to a high enough output current to drive the BC550. It works fine as posted. However, I'm wondering if there isn't a more elegant way to drive the BC500 using only one stage of the TL072, letting me use the other stage for something else, say, variable output power or a filter?
3) I have read about boosting the output current of opamps with an emitter follower in the feedback loop, but I don't quite get how to fit it in. I have done some searching but I most of the sites I find are class exercises which aren't helping much to a spare-time hobbyist like me.

Any thoughts? Why is the opamp working at all? Can I get a more elegant single-stage current drive for the input to the 8051? Any glaring errors in my work?


Thanks for any thoughts!

Torben

 

[mvs sarma]

Hi Torben,

what is the purpose of R7 from base of bc5xx to Vcc.
you are trying to amplify the detected +ve peaks of the signal . thus the diode D1 from base to earth is OK . Perhaps you need to remove the bias for the final transistor,

After that try reducing the initial 22k in the input to say 10K, bypass the 2nd stage and try reaching the base of BC5XX

All the best

 

[JimB]

drives a 38,400Hz square wave through two opposing pins to a 40kHz ultrasonic piezo transducer.

Any glaring errors in my work?

Yes, 38.4kHz is not 40khz. The transducers have a very sharp resonance at 40 khz, and 38.4khz is just not good enough.

A quick and dirty test on the pair of transducers I have here shows the 3dB bandwidth to be 39.4 to 40.9kHz, and at 38.4khz the response was 13.5dB down.
Not conducive to good operation.
I suggest you adjust your frequency to 40 khz.

JimB

 

[Nigel Goodwin]

Transducers aren't that accurate - don't adjust to 40KHz, adjust for maximum output - which 'may' be 40KHz, but may well be not. This is why they usually come as matched pairs! - don't mix up different transmitters and receivers.

 

[Roff]

The op amp stages are for voltage gain, not current drive. If you don't understand how feedback around an op amp sets voltage gain, Google some op amp tutorials.

 

[Torben]

Yes, 38.4kHz is not 40khz. The transducers have a very sharp resonance at 40 khz, and 38.4khz is just not good enough.

A quick and dirty test on the pair of transducers I have here shows the 3dB bandwidth to be 39.4 to 40.9kHz, and at 38.4khz the response was 13.5dB down.
Not conducive to good operation.
I suggest you adjust your frequency to 40 khz.

JimB

Hi Jim,

Thanks for the hint, but I am not too worried about the piezo response at the moment. As I noted, the thing does actually work fine, but right now I just want to know more about opamp current boost--not to increase range (when I get to that point I'll probably try to put the transmitting piezo closer to 40kHz) but to free up half the TL072. The 38.4kHz just works more easily for the moment with the crystal I'm using, which I selected for easy serial communications setup.

But yeah, that 13.5 dB down you found is certainly not something I would like in the final circuit.


Thanks again,

Torben

 

[Torben]

The op amp stages are for voltage gain, not current drive. If you don't understand how feedback around an op amp sets voltage gain, Google some op amp tutorials.

I do (mostly) understand feedback loops, at least the basics. I also understand that opamps in general are voltage amps and not current amps and have low output current and output impedance. What I seem to be having trouble with is understanding why an emitter follower in the feedback loop should help with current output, so I'm having trouble visualizing how it fits together. And yes, it does seem like it should be fairly obvious but I'm stumbling on it for some reason.

Anyway, I'll keep trying today and if I don't get it by tonight I'll post again.


Torben

 

[audioguru]

The opamps are connected as inverting amplifiers with a fairly low input impedance. I think the piezo transducer needs an amplifier with a high input impedance which would be a non-inverting amplifier.

The original circuit uses an LM358 that has very poor high frequency response but its minimum supply voltage is 3V.
The TL072 doen't meet its spec's when its supply voltage is lower than 7V.

Use an MC34071 single, MC34072 dual or MC34074 quad high speed opamp that also has a minimum supply voltage of only 3V.

In your circuit the first opamp has a voltage gain of 21.4 and the second opamp has a voltage gain of 4.7 for a total gain of 100.6.
A single MC3407x opamp can do it by itself at 40kHz.

Opamp current boost? Why?
An opamp has a max output current of about 20mA. The BC550 transistor needs an input voltage of 0.7V and an input current of only 100uA.

 

[Torben]

Hi Torben,

what is the purpose of R7 from base of bc5xx to Vcc.

It's just the remains of a biasing voltage divider for the BC550; there used to be a pot going to ground and its wiper went to the BC550. It didn't appear to affect things too much so I removed it--but I'll try putting it back and see what kinds of results I get.

you are trying to amplify the detected +ve peaks of the signal . thus the diode D1 from base to earth is OK . Perhaps you need to remove the bias for the final transistor,

After that try reducing the initial 22k in the input to say 10K, bypass the 2nd stage and try reaching the base of BC5XX

All the best

Thanks! I'll give that a try. I'll also try with another opamp--which is rated for 5V supply or less --and see what happens.


Thanks for the hints,

Torben

 

[Torben]

The opamps are connected as inverting amplifiers with a fairly low input impedance. I think the piezo transducer needs an amplifier with a high input impedance which would be a non-inverting amplifier.

Ah, OK. My bat detector uses a transistor preamp instead of an opamp, so maybe I'll give a version of that a shot too. But I'll definitely try the non-inverting configuration and see what I can get out of it.

The original circuit uses an LM358 that has very poor high frequency response but its minimum supply voltage is 3V.
The TL072 doen't meet its spec's when its supply voltage is lower than 7V.

Makes sense. Might explain the apparently quite low current it's able to deliver. I was surprised to see it working at all, but working out of spec makes sense.

Use an MC34071 single, MC34072 dual or MC34074 quad high speed opamp that also has a minimum supply voltage of only 3V.

In your circuit the first opamp has a voltage gain of 21.4 and the second opamp has a voltage gain of 4.7 for a total gain of 100.6.
A single MC3407x opamp can do it by itself at 40kHz.

OK. I've been wanting an excuse to make an order from DigiKey anyway.

Opamp current boost? Why?
An opamp has a max output current of about 20mA. The BC550 transistor needs an input voltage of 0.7V and an input current of only 100uA.

If it's not necessary--e.g. if I set things up correctly so that one section of the opamp can handle the task alone--then that's even better. I was just hoping to not have to use the whole TL072 when from what I (think I) understand it shouldn't be necessary.

I'll try your hints and see what I can come up with.


Thanks,

Torben

 

[Torben]

Transducers aren't that accurate - don't adjust to 40KHz, adjust for maximum output - which 'may' be 40KHz, but may well be not. This is why they usually come as matched pairs! - don't mix up different transmitters and receivers.

Thanks Nigel! I don't really have the gear to test this pair myself, but they came in the same little ziplock baggie as a pair so I can only hope they're matched. The datasheet does show them down about 6-7 dB at 38.4 kHz: https://www.electro-tech-online.com/custompdfs/2007/10/t400s16.pdf so I figure I'm wasting power compensating for not matching their resonant frequency, but for the moment I'm OK with that. This a learning project, mostly.


Torben

 

[Nigel Goodwin]

Thanks Nigel! I don't really have the gear to test this pair myself, but they came in the same little ziplock baggie as a pair so I can only hope they're matched. The datasheet does show them down about 6-7 dB at 38.4 kHz: https://www.electro-tech-online.com/custompdfs/2007/10/t400s16-1.pdf so I figure I'm wasting power compensating for not matching their resonant frequency, but for the moment I'm OK with that. This a learning project, mostly.

If they were supplied in the same bag, it's reasonable to expect they are a matched pair - I presume they are marked TX and RX?, as they are completely different.

 

[Torben]

If they were supplied in the same bag, it's reasonable to expect they are a matched pair - I presume they are marked TX and RX?, as they are completely different.

That's my hope! Yes, one is marked 400ST and the other 400SR, which matches what the datasheet says.


Torben

 

[audioguru]

The ultrasonic receiver transducer drops its level only 2dB when it has a 10k load so an inverting opamp amplifier with a 10k input resistor is fine.
EDITED.

 

[Nigel Goodwin]

The IR receiver transducer drops its level only 2dB when it has a 10k load so an inverting opamp amplifier with a 10k input resistor is fine.

Is it just me?, or haven't we been talking about ultrasonic transducers?.

 

[Roff]

Is it just me?, or haven't we been talking about ultrasonic transducers?.

No, it's not just you. I had the same thought, but restrained myself.

 

[Torben]

Is it just me?, or haven't we been talking about ultrasonic transducers?.

No, you've clearly gone barking mad. Just kidding. I figured it was just a typo as IIRC audioguru was also posting in another thread about IR detectors.


Torben

 

[audioguru]

Is it just me?, or haven't we been talking about ultrasonic transducers?.

OOPs. Too many IR posts lately. I edited my reply about ultrasonic transducers.

 

[Torben]

The opamps are connected as inverting amplifiers with a fairly low input impedance. I think the piezo transducer needs an amplifier with a high input impedance which would be a non-inverting amplifier.

The original circuit uses an LM358 that has very poor high frequency response but its minimum supply voltage is 3V.
The TL072 doen't meet its spec's when its supply voltage is lower than 7V.

Use an MC34071 single, MC34072 dual or MC34074 quad high speed opamp that also has a minimum supply voltage of only 3V.

OK, I have looked up the MC3407x and it looks like they're discontinued. A suggested replacement is the TLE214x, which seems to fit the bill, except it doesn't have FET inputs so I might need to go to the non-inverting amp after all (not sure about this yet).

However, I got a spice model for the TLE2141 and with some changes in my circuit it works nicely--in my simulation. Haven't ordered any yet. The TL072, which does work at 5V when using both stages, totally fails when using one stage at 5V. The TLE2141 does the trick.

Some other things I would like to clarify:

1) I think that the voltage on the base of Q1 does not suffer limiting from D1 because of the base-emitter diode action of Q1. Is this true?
2) I think I'm OK pulling the bias on the positive input of U1 so hard to Vcc because I *want* to slam the output from 5V to 0V when receiving. Is this reasonable?

Hoping to order some TLE2141s soon if this is a reasonable circuit.


Thoughts? Thanks!

Torben

 

[Roff]

You have biased the input at 4.5V, when the input common mode range extends to only 3V. You should probably set the input bias voltage at about 2V.
You have a highpass filter (220pF, 10k) on the input at 72kHz. Your input capacitor should be at least 1.6nF, which would set the corner at 10kHz.
EDIT: I see why you picked the values you did. In the simulation, it works with them, and it doesn't with the values I recommended.
I'm not sure I would trust the spice model. You are clearly exceeding the common mode range by a large amount. You can probably get the circuit to work, but you should probably be prepared to juggle component values when you build hardware.