Yes , or No ?

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Yes , or No ?

Post by Bernius1 » Thu Feb 24, 2005 4:54 am

OK, I'm having a Retardo Mentalbloc day, so I'll ask a " yes or no " question. If I get enough 'yeas', I'll pursue the project. Here goes;
Two (2) 2V triangular waves, 90° out of phase, have an absolute value difference of |2V| at all times (ideal). SO, can a single op-amp (or comparator ) act as an absolute value detector for TWO inputs ? My hope is to use the feedback of the detecting op-amp with an error amp to create an accurate two-pole time base.
Can't we end all posts with a comical quip?

dyarker
Posts: 1752
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Re: Yes , or No ?

Post by dyarker » Thu Feb 24, 2005 6:00 am

Sorry, I can't help with the triangle waves. My prescription for "Retardo Mentalbloc" is 4oz. alcohol, go to bed 30 minutes earlier than normal (not more), wake up normal time, 1 cup coffee, hot shower, 2nd cup coffee; brain is now "in gear". :D<p>[ February 24, 2005: Message edited by: Dale Y ]</p>
Dale Y

Ron H
Posts: 360
Joined: Tue Dec 18, 2001 1:01 am
Location: Boise, ID
Contact:

Re: Yes , or No ?

Post by Ron H » Thu Feb 24, 2005 11:23 am

I'm baffaloed by your question. How can two waveforms, 90 degrees out of phase, have a constant absolute value difference of 2 volts when they cross each other (diff=0) twice per cycle? Am I missing something?

josmith
Posts: 340
Joined: Tue Dec 04, 2001 1:01 am
Contact:

Re: Yes , or No ?

Post by josmith » Thu Feb 24, 2005 1:55 pm

And how does it become a time base?
Try Ginko
No!

User avatar
haklesup
Posts: 3046
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Yes , or No ?

Post by haklesup » Thu Feb 24, 2005 3:12 pm

Without fully understanding the application.<p>Can you not use the principal of superposition to simply add the two waveforms using a voltage divider (one wave at the top and one at the bottom) and take the error signal from the summing node (in the middle). You can now take that error signal and apply it to a comparitor to drive an output when the error is exceeded. I imagine that the two resistors would have the same value and that value would be dictated by the current you wanted at the output and how much was available at the inputs.<p>You could also place the waveforms at the inputs to an op amp and with the proper gain as set by the feedback R you would obtain a constant output as long as the difference between the two inputs is constant. This may be called a difference amp or summing amp. You may also need to invert one of the signals at the input.<p>I also do not see what you mean by two pole time base. DO you mean a two-phase clock?

dyarker
Posts: 1752
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Re: Yes , or No ?

Post by dyarker » Thu Feb 24, 2005 11:04 pm

The way I see it, from 0° to 90° the difference is 2V, then the difference decreases to 0V at 135° (where both triangle waves are 1V). From 180° to 270° the difference is 2V. Then another cross-over at 315° at -1V.<p>Accuracy is going to depend on source of triangles, and be limited by the analog approach. (Of course I don't know your end goal either.) I'm more digital oriented, I hear "time base" and "90°"; and I automatically jump to times 2 clock, 2 flip-flops and a NAND gate.<p>(took my own advice, I'm half way through 3rd cup of coffee ;)
Dale Y

Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Re: Yes , or No ?

Post by Bernius1 » Fri Feb 25, 2005 4:07 am

Hacklesup; You're right, it's a 2-phase clock. At 0°, 'a' is 2V & 'b' is 0V ; At 45°, they're both at 1V, so the sum is 2V; At 90°, 'b' is 2V & 'a' is 0V; at 135° , they're +1 & -1 with an absolute sum of 2. Like a flip-flop with Q and 'inverted-Q' (how the he** do you get the A or Q with the Macron atop ?) outputs. But unlike sine waves, the absolute value between the waves remains 2V ( negating noise ). So the absolute value amp and error amp are not to correct either wave alone, but to maintain the |2V| sum.
It's gonna turn out to be a puzzle in edge-triggered logic.
BTW? Isn't the tech's best friend named Folgers ?
Can't we end all posts with a comical quip?

Bernius1
Posts: 349
Joined: Sat Sep 21, 2002 1:01 am
Location: NY
Contact:

Re: Yes , or No ?

Post by Bernius1 » Fri Feb 25, 2005 5:02 am

Here's one idea, but it's single pole.
get the PDF version!
Can't we end all posts with a comical quip?

Will
Posts: 310
Joined: Tue Feb 11, 2003 1:01 am
Location: Katy Texas
Contact:

Re: Yes , or No ?

Post by Will » Fri Feb 25, 2005 10:53 am

As I see it - if you connect the two voltage sources through two resistors in summing mode to the negative terminal of an op amp then size the feedback resistor so that, for the sum, it produces an output wave form which alternates between - (minus) 2 volts to plus 2 volts the the wave form will look like this - - (minus) 90 degrees - minus 2 volts then a linear ramp to plus two volts at zero degrees, then a flat plus 2 volts from zero to 90 degrees then, from 90 degrees to 180 degrees a linear ramp down to minus 2 volts, then a flat 2minus 2 volts from 180 to 270 volts then a linear ramp from minus 2 volts to plus 2 volts from 270 degrees to 450 degrees. The period of the waveform (Zer volts to zero volts) is from minus 45 degrees to 315 degrees - sense for a positive voltage anywhere in the minus 45 to plus 135 degree space and you have a clock of the correct frequency ?
BB

Will
Posts: 310
Joined: Tue Feb 11, 2003 1:01 am
Location: Katy Texas
Contact:

Re: Yes , or No ?

Post by Will » Fri Feb 25, 2005 10:55 am

Sorry ! - I forgot to add a humorous or interesting quote - In the words of the Scottish bard Robbie Burns = Is it wise to remain silent a be thought a fool ? Or to speak up and dispel all doubt ?
BB

Post Reply

Who is online

Users browsing this forum: No registered users and 27 guests