MiniPulse Pluse audio options

[Tibuck19]

I thought i would start a new post for the Mini Pulse plus Audio alternatives.

Before we start off i would like to quote a posting from Carl Moreland
" Welcome to the new Tech forum

The purpose of this forum is the exchange of ideas relating to the design and operational theories of metal detectors, magnetometers, earth resistivity devices, and any other geophysical instruments that are used in the field of treasure hunting. I hope we can have some interesting discussions and I urge everyone to participate regardless of experience. There is no question too basic for this forum."
Reference posting : http://www.geotech1.com/forums/showthread.php?7033-Welcome!

I do not know how it resonates for each of you, However, my interpretation is you do not need to be a IC chip designer to add to knowledge base, Can we not all agree, we are here for same reasons, We all have a shared passion for metal detectors, and electronics and communicating ideas with others around the world!

Now back to the idea's at hand,

Godigit1,Geoscash1 and a few others are working on a GB Fantastic
Teleno and a few others are working on improving the Preamp Bring the mmp one step closer to a gold detector!
Your all doing a great job! i don't understand half of what you guys are doing, but its great!

So now for the Audio, Please correct me if I am wrong, The audio out seems to have a hash saw tooth wave form, and what we might want is more of a soft triangle wave form. So what do we need to do to make that happen ? Maybe adding an addon board to PL5 have it go in too a cem3340 or some other VOC chip, or does the change need to happen before or after U10.

[Tibuck19]

Well maybe i was wrong, Perhaps i should put in 3 years of electronic engineering schooling and come back then.

[waltr]

So now for the Audio, Please correct me if I am wrong, The audio out seems to have a hash saw tooth wave form, and what we might want is more of a soft triangle wave form. So what do we need to do to make that happen ? Maybe adding an addon board to PL5 have it go in too a cem3340 or some other VOC chip, or does the change need to happen before or after U10.

An RC will round off the saw tooth if you only wish to have a less harsh sounding output.

If I am looking at the correct schematics it uses a 555 chip to output audio which is then RCR filtered (680R-470n-680R) which will round off the 555's Square wave output.

What does this look like on a scope and what does it sound like to you?

On my HammerHead the audio is a square wave and filtered like the MPP and sound fine to me.
However, I do a VCO (using a PIC 12F1501) to have varying tones (see my HH2 thread for full info).

There are also many other audio circuits in the forum that are nice additions/mods to any of the simpler PI circuits.

[dbanner]

I stumbled across a fantastic audio circuit used in the cscope 990b/1220b metal detectors. It uses one dual opamp and a dflop (4013), now only if I can figure out how to adapt it to a pi machine. I'll have to do some analysis on just what's going on with that circuit.

[6666]

I stumbled across a fantastic audio circuit used in the cscope 990b/1220b metal detectors. It uses one dual opamp and a dflop (4013), now only if I can figure out how to adapt it to a pi machine. I'll have to do some analysis on just what's going on with that circuit.

Are you still working on this ?

[dbanner]

I tried to model the cscope circuit with ltspice. I just couldn't get it to oscillate, no matter what I tried.
I began to suspect that the original schematic contained some superficial error. But it is correct. (I traced it from photos of the PCB). The problem is likely that I wasn't modelling it correctly.
I never pursued it any further.
The real question is whether or not the circuit is a true VCO, ie, does it run off of a control voltage. Or does it run off of a pulse train of varying amplitude.
The latter would be inappropriate for cut and paste adaptation to a pI.
The circuit is interesting though, and I'm sure that re-engineering the concept as a basis of a VCO for a pI detector using the 4013 chip is a rational goal.

[Qiaozhi]

I tried to model the cscope circuit with ltspice. I just couldn't get it to oscillate, no matter what I tried.
I began to suspect that the original schematic contained some superficial error. But it is correct. (I traced it from photos of the PCB). The problem is likely that I wasn't modelling it correctly.
I never pursued it any further.
The real question is whether or not the circuit is a true VCO, ie, does it run off of a control voltage. Or does it run off of a pulse train of varying amplitude.
The latter would be inappropriate for cut and paste adaptation to a pI.
The circuit is interesting though, and I'm sure that re-engineering the concept as a basis of a VCO for a pI detector using the 4013 chip is a rational goal.

I tried simulating the circuit in LTspice, and it worked first time.
Simulation attached.

[dbanner]

Ah. Thanks so much for this.

[dbanner]

Now that I am in quarantine, I have been feeling motivated once again to re-engage in hobby building and tinkering.
Only one big problem..... I AM RUNNING OUT OF SOLDER!!!!

All local non-essential businesses are shut down. Supermarket does not sell solder.
Rrrrrr.....

[dbanner]

I tried simulating the circuit in LTspice, and it worked first time.
Simulation attached.

I ran your simulation. It works very well.

I noted that the control voltage (VIN) must be positive only.

.
I'm not sure what voltage is present at the minipulse output. Maybe with a little tinkering this circuit might be adapted to work as a nice little VCO ad-on. If any diy pulse detector can be made to sound like the Cscope 1220B, that's got to be a good thing.

How would I go about plotting the voltage to frequency response of this circuit in LTspice, does anyone know? What parameters for VIN should I use? Should I define the voltage source using PWL?

I guess I'd be aiming for a triangle shape waveform for the control voltage with linear rise and fall.

[waltr]

I ran your simulation. It works very well.

I noted that the control voltage (VIN) must be positive only.

.
I'm not sure what voltage is present at the minipulse output. Maybe with a little tinkering this circuit might be adapted to work as a nice little VCO ad-on. If any diy pulse detector can be made to sound like the Cscope 1220B, that's got to be a good thing.

How would I go about plotting the voltage to frequency response of this circuit in LTspice, does anyone know? What parameters for VIN should I use? Should I define the voltage source using PWL?

I guess I'd be aiming for a triangle shape waveform for the control voltage with linear rise and fall.

Add another Voltage Source and connect to VIN & ground.Mouse over and right click and then ADVANCED to set parameters.
Select PULSE. Set to RISE TIME to the period and Fall time very short to produce a Ramping Voltage (saw tooth).
A Triangle is done the same way except rise and fall times are the same and half the period.

[dbanner]

Add another Voltage Source and connect to VIN & ground.Mouse over and right click and then ADVANCED to set parameters.
Select PULSE. Set to RISE TIME to the period and Fall time very short to produce a Ramping Voltage (saw tooth).
A Triangle is done the same way except rise and fall times are the same and half the period.

Thanks waltr, I'm going to have a play with the Pulse parameters in the ADVANCED Vsource settings. Should be very easy to set up.
Will post a screenshot of the results. I may tinker with the values of the circuit to see if I can get a fairly linear response, or at least to get a feel for what the limits of the circuit are.

[Qiaozhi]

I suspect what you're really trying to do is to generate a plot of frequency versus control voltage.
To do that you need to use the .measure facility in LTspice. Please see attachments.

According to the information I have on the CS-1220, you need to adjust the trimmer R71 (47k) in the schematic [R6 in the simulation] so that the audio frequency is 1060Hz when there's 700mV at the control voltage input. By a matter of trial and error, I found that setting R6 to 14k06 gives an audio frequency of 1060.91Hz, which is close enough.
This trimmer is obviously setting the maximum audio frequency. Then I set up a parametric sweep of the control voltage from 10mV to 700mV in steps of 10mV.
The .measure command determines the time at which the audio output voltage first crosses 500mV after a delay of 30ms, and also where it crosses 500mV the third time. The difference will be the period of the audio signal.

To plot the period versus vc, run the simulation (70 runs), and enter <CTRL> L. This displays the log file with the measured results. Click the right-hand mouse button in the log window and select "Plot .step'ed .meas data". This will display a plot of the period versus control voltage. To display the frequency versus vc, right click on the waveform name (period) and change the expression to 1/period.

I did try to add a second .measure command: ".meas tran freq param 1/period" to automatically do this last calculation, which displays correctly in the log file, but results in a blank plot window. Don't know what's going on there.

[Qiaozhi]

I did try to add a second .measure command: ".meas tran freq param 1/period" to automatically do this last calculation, which displays correctly in the log file, but results in a blank plot window. Don't know what's going on there.

OK - I've figured it out.
You need to right click in the [blank] plot window and select View>Visible Traces. Then it lists both period and freq.
All you need to do is add ".meas tran freq param 1/period" into the simulation.

Simple when you know how.

[waltr]

Thanks Q,
Good stuff.