Heart Beat Counter Circuit

On a sunny day (10 Dec 2006 09:28:02 -0800) it happened snipped-for-privacy@gmail.com wrote in :

Use Microchip PIC microcontroller, you will have to proram it, learn ASM. (or other micro processor) every other way probably needs more hardware.

Oops- forgot to mention that its all in hardware- we're not allowed to use software for this project!

Jan Panteltje wrote:

(...)

Is there room for a 3 bit counter on your board? You could increment it during each 15 second interval and it would go to 0x100 at the end of

60 seconds. When the MSB goes high, reset both counters.

Look up 'seven segment decoder'.

--Winston

On a sunny day (10 Dec 2006 09:50:21 -0800) it happened snipped-for-privacy@gmail.com wrote in :

Ok, homework, counter:

3 seven segment displays. 3 BCD to seven segment decoders / drivers 3 quad latches 3 74HC[T]90 decade counters with BCD output in series to count pulses. 1 555 timer running at some high frequency 1 binary divider so you get the last Q output high after 60 seconds, this then triggers some one-shots that 1) latch the decade counter output, 2) reset the decade counters so the display is updated. That was the part you did not know? For better accuracy you could use a crystal (not a 555 timer). All is available in 74HC range for 5V.

you didnt spec accuracy :-) rough idea

555, at a 15 sec count binary up counter 74 or CD4000 series 74hc138 will decode the lowest 3 lines of the counter into induvidual lines

Thats a start, what would you do next?

martin

Yes.

1 CPLD.

Yes.

included in CPLD.

I have a few silly questions- I'm not too familar with all of this stuff soo....

-are you saying I should use three decade counters together? how does one set them up to count together?

- what is a binary divider? What is "Q" output? Would I set the "Q" output to the reset terminal of the coutners to that they reset every

60 seconds?

- what are quad latches?

Sorry if I'm painfully novice at this! I'm just trying to imagine the actual set up of this thing on our board and am not sure what some of these components are!

Thanks again, Rachel

Le> On a sunny day (10 Dec 2006 09:50:21 -0800) it happened snipped-for-privacy@gmail.com

But how exactly does one reset the counters? Is it just a matter of sending that pulse every 60 seconds to a reset terminal? Also, do you have any suggestions for specific counters to use?

Thanks again!

W>

--
I\'ll be happy to help you, but I think you have a problem with that
15 second measurement in that you\'ll have to multiply it by four to
get the one minute rate and you\'ll lose some data in the units or
possibly even tens of beats per minute doing it that way.  If the
design rules aren\'t fixed, I suggest that you count beats for a full
minute, then latch and display that data at the same time that you
reset the counter and start another count cycle.  That way you\'ll
have new data coming in every minute and the counter will always be
working in the background.  If you\'re interested, I can post a
schematic to alt.binaries.schematics.electronic, or email you one
since you can\'t access abse from Google.

That would be awesome! The problem is that we really do have to count for only 15 seconds and then multiply by 4 to get the beats per minute. My teacher suggested doing this by moving some of the wires over two places(?) in the counter because that would multiply everything by four. I'm not sure I truly understood his hint. I would love to post the 2 schematics I have right now so you could see what I'm working on- do you know where/how I could post these and would you even be willing to take a look at them?

Thanks!

Rachel

John Fields wrote:

--- Sure. Post them to a web page somewhere or, if you can't do that, email them to me and I'll post them to abse so everybody can play.

That multiply-by-four is going to make things nasty, and there's no need for it.

Here's what your instructor was talking about:

Lets say you measure beats for 15 seconds using an 8-bit binary counter and you get:

0000 1111

which converts to decimal 15.

Shift the pattern one bit to the left:

0001 1110

and that becomes decimal 30.

Shift it once more and it becomes:

0011 1100

which is decimal 60, but it's been done in binary and it'll have to be converted to Binary Coded Decimal (BCD) in order to be displayed using easily available parts.

If it's your instructor's intention to have you learn to do binary multiplication as well as binary-to-BCD conversion then, for your application, doing the conversion would probably be easiest by counting the beats for 15 seconds using a binary up/down counter in 'up' mode, wiring the Q outputs to the D inputs so that Q0 goes to D2, Q1 goes to D3, Q2 goes to D4, and so on. Then, when the 15 second accumulation period ends, do a load. That will load a number

4 times greater than what the counter counted, back into itself.

Also, when the 15 second period ends, clear a BCD counter.

Now, reverse the counting direction of the binary counter and use a convenient clock to simultaneously clock both the binary and the BCD counter. Make it so that when the binary count gets to zero the Q outputs of the BCD counter are loaded into BCD-to-seven-segment decoder-drivers, and that'll cause the displays to show four times the number of beats picked up in the previous 15 seconds.

The conversion and display can be done in microseconds so, unless the one-reading-per-minute spec is cast in stone, for extra credit you might want to think about doing more than just one measure-convert-display cycle per minute.

-- JF

Cool. I just sent you an email with my current-state schematics. Thanks for the description of the multiply by 4 hint- I totally get it now. I also think that I understand theoretically how this could work now, but still dont get how the BDC and the counter would actually fit into the circuit? Just to clarify, is the counter just a series of JK flip flops? Hmm...

R.

John Fields wrote:

Please post at the bottom of messages, not at the top. Makes it easier to read.

You can reset counters via a pulse on the reset line. Make sure to use counters with unconditional reset, not some where the reset will only happen upon the next clock.

--
Regards, Joerg

http://www.analogconsultants.com

What school are you attending where they teach you how to do outdated designs??? Just glad the teacher didn't insist on you doing it all with tubes and relays! Really, get with it dude and learn why we do all this with microcontrollers today.

Luhan

haha, I KNOW! Not to mention, the teacher gave us this project just recently and expects us to complete it all during exam week!

Rachel

Luhan wrote:

I note you're not using a microcontroller, I'll assume there's a reason for that.

some hints, look into using counters that can drive the display directly, or maybe BCD counters and BCD-TO-7segent converter chips.

at the end of the 45 second period you need to send a reset pulse to the counters. a R-C high-pass filter could be one way to do that.

Bye. Jasen

have you looked at the data sheet for these parts?

divide-by-two counter, eg: flip-flop with D wired to Q-bar

What is "Q" output?

non-inverting

Would I set the "Q"

google for a datasheet.

Bye. Jasen

Chill! slow rate logic is good for teaching, you can debug it with a bunch of LEDs.

Sure a micro coild solve this task with more accuracy and less hardware, but it doesn't teach clocks, gates, counters, and latches.

Bye. Jasen

They got you using hardware logic to do heart monitoring while I'm using micros to do a yoyo.

Luhan :)