DTL (diode AND gate)

G

George Herold

Guest
Hi all, Another digital question.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch pulling one bit high.

The boss wanted a finer division, which I first thought would take
AND gates, but then I thought about just diodes to only pull
high those lines I want. (Rather than explain more here's a circuit
fragment.)

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

So this gives me
2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?

TIA

George H.
 
On Friday, February 26, 2016 at 7:47:53 AM UTC-8, George Herold wrote:
..
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch
The boss wanted a finer division, [add diode gates]
So this gives me
2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?
Click to expand...

Should work OK. The shown values will work with CMOS if you don't try
to operate below 3V power (you lose some logic margin by diode drops).

The select circuit is now more complex to wire, and harder to understand,
than a three digit thumbwheel switch...
 
On Friday, February 26, 2016 at 3:56:50 PM UTC-5, whit3rd wrote:
On Friday, February 26, 2016 at 7:47:53 AM UTC-8, George Herold wrote:
.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch
The boss wanted a finer division, [add diode gates]
So this gives me
2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?

Should work OK. The shown values will work with CMOS if you don't try
to operate below 3V power (you lose some logic margin by diode drops).

The select circuit is now more complex to wire, and harder to understand,
than a three digit thumbwheel switch...
Click to expand...

Grin.... (thanks whit3rd.)

George H.
 
On Friday, February 26, 2016 at 7:47:53 AM UTC-8, George Herold wrote:

I'm doing this pulse thing.

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0
Click to expand...

Am I missing something, or does the RCO output on IC14 give you
the end-of-time output you want? What's with the NAND gates?
 
On Friday, February 26, 2016 at 4:34:12 PM UTC-5, whit3rd wrote:
On Friday, February 26, 2016 at 7:47:53 AM UTC-8, George Herold wrote:

I'm doing this pulse thing.

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

Am I missing something, or does the RCO output on IC14 give you
the end-of-time output you want? What's with the NAND gates?
Click to expand...

Oh, this is just a fragment, I can post the whole thing on Monday,
and you can rip it apart. I set a bit and then down count... all the
RCO's go to an RS and then into a D-flip which shuts things down on the next clock cycle. (My digital design is a bit "organic". I made something
that mostly worked and then cleaned up the bit's that didn't...
There is most likely some better way to do it.)

George H.
 
On Friday, February 26, 2016 at 5:05:33 PM UTC-5, Tim Wescott wrote:
On Fri, 26 Feb 2016 07:47:46 -0800, George Herold wrote:

Hi all, Another digital question.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch pulling one bit high.

The boss wanted a finer division, which I first thought would take AND
gates, but then I thought about just diodes to only pull high those
lines I want. (Rather than explain more here's a circuit fragment.)

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

So this gives me 2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?

Check to see that you have enough margin. Margin can be improved by
using schottkey diodes and/or 74HTC161 counters.

Check where pins 8 and 9 of U$8 are going.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Click to expand...

Cmos and 5V supply. The pins look ok... I did all the odd numbers first
they go straight. Anyway I don't think we need finer divisions in time.
(That's my bosses idea.)
The whole circuit sends a current pulse to a resistor* and measures the
voltage. (to determine the energy delivered.) I've got two knobs,
one for T and one for I... I(current) should have the finer divisions.

I did the analog part ~1 year ago, but now I need to pick
the current sense resistors. I did a series string today,
but I don't like it.. too much heat, so I'll try parallel.
For what ever reason it's harder for me to think about
parallel, conductance rather than resistance.

George H.
*at the cost of another wire a FET would be better.
 
On Fri, 26 Feb 2016 07:47:46 -0800, George Herold wrote:

Hi all, Another digital question.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch pulling one bit high.

The boss wanted a finer division, which I first thought would take AND
gates, but then I thought about just diodes to only pull high those
lines I want. (Rather than explain more here's a circuit fragment.)

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

So this gives me 2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?
Click to expand...

Check to see that you have enough margin. Margin can be improved by
using schottkey diodes and/or 74HTC161 counters.

Check where pins 8 and 9 of U$8 are going.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
 
On Fri, 26 Feb 2016 16:57:18 -0800, George Herold wrote:

On Friday, February 26, 2016 at 5:05:33 PM UTC-5, Tim Wescott wrote:
On Fri, 26 Feb 2016 07:47:46 -0800, George Herold wrote:

Hi all, Another digital question.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch pulling one bit high.

The boss wanted a finer division, which I first thought would take
AND gates, but then I thought about just diodes to only pull high
those lines I want. (Rather than explain more here's a circuit
fragment.)

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

So this gives me 2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?

Check to see that you have enough margin. Margin can be improved by
using schottkey diodes and/or 74HTC161 counters.

Check where pins 8 and 9 of U$8 are going.

--

Tim Wescott Wescott Design Services http://www.wescottdesign.com

Cmos and 5V supply. The pins look ok... I did all the odd numbers first
they go straight. Anyway I don't think we need finer divisions in time.
(That's my bosses idea.)
The whole circuit sends a current pulse to a resistor* and measures the
voltage. (to determine the energy delivered.) I've got two knobs,
one for T and one for I... I(current) should have the finer divisions.

I did the analog part ~1 year ago, but now I need to pick the current
sense resistors. I did a series string today,
but I don't like it.. too much heat, so I'll try parallel.
For what ever reason it's harder for me to think about parallel,
conductance rather than resistance.

George H.
*at the cost of another wire a FET would be better.
Click to expand...

Look closer. You've got a wire going from pin 8 and a diode going to
that same wire from pin 9. So if pin 9 is high and 8 low, you'll have an
effective short.



--
www.wescottdesign.com
 
On Saturday, February 27, 2016 at 11:51:38 AM UTC-5, Tim Wescott wrote:
On Fri, 26 Feb 2016 16:57:18 -0800, George Herold wrote:

On Friday, February 26, 2016 at 5:05:33 PM UTC-5, Tim Wescott wrote:
On Fri, 26 Feb 2016 07:47:46 -0800, George Herold wrote:

Hi all, Another digital question.
I'm doing this pulse thing. Pulse widths were,
1,2,4,8,16.. simple set by a rotary switch pulling one bit high.

The boss wanted a finer division, which I first thought would take
AND gates, but then I thought about just diodes to only pull high
those lines I want. (Rather than explain more here's a circuit
fragment.)

https://www.dropbox.com/s/9m05ssjm962ma0f/DTL.jpg?dl=0

So this gives me 2,3,4,6,8,12,16,24......

This seems to work just fine... is there some gotcha I'm missing?

Check to see that you have enough margin. Margin can be improved by
using schottkey diodes and/or 74HTC161 counters.

Check where pins 8 and 9 of U$8 are going.

--

Tim Wescott Wescott Design Services http://www.wescottdesign.com

Cmos and 5V supply. The pins look ok... I did all the odd numbers first
they go straight. Anyway I don't think we need finer divisions in time.
(That's my bosses idea.)
The whole circuit sends a current pulse to a resistor* and measures the
voltage. (to determine the energy delivered.) I've got two knobs,
one for T and one for I... I(current) should have the finer divisions.

I did the analog part ~1 year ago, but now I need to pick the current
sense resistors. I did a series string today,
but I don't like it.. too much heat, so I'll try parallel.
For what ever reason it's harder for me to think about parallel,
conductance rather than resistance.

George H.
*at the cost of another wire a FET would be better.

Look closer. You've got a wire going from pin 8 and a diode going to
that same wire from pin 9. So if pin 9 is high and 8 low, you'll have an
effective short.
Click to expand...

Hi Tim, OK I starred at it three times now. (It still looks OK to me...
but if I spin it this way and it's wrong I'll be sure to shout out to you.)

The U$8 is the rotary switch.. (Yeah I just made the model for it and I should
have given it a better designator... lazy on my part.) The switch is BBM,
so there should not be any shorts.

George H.
--
www.wescottdesign.com
Click to expand...
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top