View previous topic :: View next topic |
Author |
Message |
manoel_filho Guest
|
AC/DC CONVERTER OR PEAK MEASURE ROUTINE. |
Posted: Thu May 04, 2006 4:50 am |
|
|
HI.
I made a complex software to manages the AC value, basic its a ac voltmeter that will work from 0 to 600V, all digital stuff is working properly, the program have about 1800 lines, the difficult is the ac/dc converter, or how to read the peak value to calculate RMS. I already tried many circuits, but until now I dons get it working. Does anybody have a ac/dc circuit that really work? And can provide it to me.
Best Regards. |
|
|
SherpaDoug
Joined: 07 Sep 2003 Posts: 1640 Location: Cape Cod Mass USA
|
|
Posted: Thu May 04, 2006 6:58 am |
|
|
Are you measuring 50 Hz or 60 Hz power lines? Do you know which or do you have to handle both? Or are you working with generators and have to deal with fluxuating frequencies? Do you have the A/D set up to read the live voltage, but you need help getting the exact peak value? Do you have a zero crossing signal?
If you have a zero crossing sync signal you can just wait 1/4 cycle and take a reading, wait 1/2 cycle more and take another reading. The difference will give AC peak to peak. The average will give any DC offset.
If you don't have a sync you need to A/D rapidly to find the peak, though this is less accurate and susceptable to noise. Do it over several cycles for best results.
Tell us more about your application. _________________ The search for better is endless. Instead simply find very good and get the job done. |
|
|
manoel_filho Guest
|
|
Posted: Fri May 05, 2006 6:57 am |
|
|
Hi.
This is the main problem. I don"t have zero crossing. I handle 60hz power lines only, and I facing difficut to set my code to verify the peak value. In fact I need two things. A AC/DC converter, I did some experiences with on rail to rail OP that seans to fit, the second thing is tha code to do that peak verification. I need to sample the ac voltage at the end of OP, verify when the voltage starts to drop and pick the peak value. I think that way the problen will be solv.
Thanks. |
|
|
SherpaDoug
Joined: 07 Sep 2003 Posts: 1640 Location: Cape Cod Mass USA
|
|
Posted: Fri May 05, 2006 7:36 am |
|
|
I see three basic approaches:
1) Software intensive - Run your A/D converter fast and use software to find the positive and negative peaks.
2) Dedicated hardware - Check out the Analog Devices AD737 chip. This is a dedicated chip to do just what you want. It costs about $3 each, but you can also get free samples if you ask them. Other companies have simillar chips.
3) If this is an educational project or a real cost sensitive high volume product you may choose to roll your own. In either case it will pay to research and really understand the circuit you choose. Start by googling "peak detector circuit" and you will get lots of good hits.
If you propose a circuit or some code we may be able to give you some pros & cons to that approach. _________________ The search for better is endless. Instead simply find very good and get the job done. |
|
|
manoel_filho Guest
|
|
Posted: Fri May 05, 2006 7:39 am |
|
|
Ok, I will take a look.
Thank you very much. |
|
|
|