Sinusoidal Phase Shift - Page 2 - Electronic Circuits Projects Diagrams Free

meastwood98 · 7th July 2009, 12:12 PM

My apologies, the circuit does work at 100Hz, 500Hz and 1kHz. However when i reduce the frequency to 10Hz, the phase shift increases. Is this the same with your simulation?

Thanks

ericgibbs · 7th July 2009, 12:39 PM

Quote:

Originally Posted by meastwood98

My apologies, the circuit does work at 100Hz, 500Hz and 1kHz. However when i reduce the frequency to 10Hz, the phase shift increases. Is this the same with your simulation?

Thanks

hi
This was your reply to my post asking for limits.

Quote:

Minimum frequency = 1Hz
Maximum frequency = 100Hz

I have not tested at 10Hz, your spec was 100Hz thru 1000Hz.

EDIT:
I see what I have done wrong, I have misread 1HZ as 1KHZ!!!, what a plonker, very sorry about that.

I will beat my self with a pain stick for 10 minutes.

ericgibbs · 7th July 2009, 12:44 PM

hi,
See my edit.

I will rework the circuit to suit 1HZ and 100Hz limits, get back to you.

ericgibbs · 7th July 2009, 01:04 PM

hi,
At these low frequencies the size of the capacitors makes this technique not a feasible method.

From about 10Hz thru 100Hz the coupling caps are 1000uF and the phase shift is only just acceptable, less than 10Hz the phase shift is not acceptable.

I'll think of an alternative solution, let you know the outcome.

EDIT:
Can you say what is the application is for this phase shifted signal, ie: whats the end result.?

meastwood98 · 7th July 2009, 01:26 PM

We have a linear sensor that gives out two sinusoidal signals 90 degrees apart, these are put through an interpolator (which we cannot play with the electronics inside) and then into a PC for Transmission Error Measurement. The problem we are facing is a large error on the PC, which is caused by the two signals not being exactly 90 degrees apart. There is approximately 2-3 degrees of error.

ericgibbs · 7th July 2009, 01:58 PM

Quote:

Originally Posted by meastwood98

We have a linear sensor that gives out two sinusoidal signals 90 degrees apart, these are put through an interpolator (which we cannot play with the electronics inside) and then into a PC for Transmission Error Measurement. The problem we are facing is a large error on the PC, which is caused by the two signals not being exactly 90 degrees apart. There is approximately 2-3 degrees of error.

hi,
Thanks for the explanation, I'm still kicking myself for misreading your posted limits.

MikeMl · 7th July 2009, 04:30 PM

Quote:

Originally Posted by meastwood98

We have a linear sensor that gives out two sinusoidal signals 90 degrees apart, these are put through an interpolator (which we cannot play with the electronics inside) and then into a PC for Transmission Error Measurement. The problem we are facing is a large error on the PC, which is caused by the two signals not being exactly 90 degrees apart. There is approximately 2-3 degrees of error.

Since you are putting the final result into a computer, presumably by using an A/D Conveter between the interpolator and the computer, why dont you just junk the interpolator, A/D both of the quadrature signals, and do your own interpolation inside the computer in software.

If linearizing the interpolation requires phase-shifting one of the quadrature signals relative to the other, this becomes a trivial operation once the signals are digitized. I'll bet that the interpolation is no more complicated than a vector multiply.

meastwood98 · 8th July 2009, 07:37 AM

The problem is that the interpolator and the computer software is something that is a permanent setup that we use with various different linear and rotarty encoders. Therefore the solution we want is to be able to just plug the sensors in as we would with an encoder without editing any software or the interpolator.

The sensors themselves only cost approx £5 and therefore cannot really be expected to give the same accuracy as an expensive encoder, however we were just interested if we could manipulate the signals to get somewhere close.

If it cant be done its not a problem, thanks anyway.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar
Title	Starter	Forum	Replies	Latest
Phase shift - Phase Angle	windozeuser	General Electronics Chat	20	20th March 2009 10:39 AM
phase shift	mayeewa	General Electronics Chat	1	27th September 2008 10:40 AM
Phase Shift	Search it	Electronic Projects Design/Ideas/Reviews	2	5th September 2008 02:13 PM
Wired up a phase shift oscillator but I dont think im getting a sinusoidal wave	Qaiphyx	General Electronics Chat	20	21st July 2008 02:37 PM
How Can I detect the phase shift between two sinusoidal signals	Abdullah2121	Electronic Projects Design/Ideas/Reviews	13	1st April 2008 04:37 AM

7th July 2009, 12:12 PM	#16
meastwood98	My apologies, the circuit does work at 100Hz, 500Hz and 1kHz. However when i reduce the frequency to 10Hz, the phase shift increases. Is this the same with your simulation? Thanks

7th July 2009, 12:44 PM	#18
ericgibbs	hi, See my edit. I will rework the circuit to suit 1HZ and 100Hz limits, get back to you. __________________ Eric " Good enough is Perfect " I will NOT answer PM's requesting technical help, please use the Forum PIC tutorials: Nigel's www.winpicprog.co.uk/ Bill's: www.blueroomelectronics.com/ Last edited by ericgibbs; 7th July 2009 at 12:48 PM.

7th July 2009, 01:04 PM	#19
ericgibbs	hi, At these low frequencies the size of the capacitors makes this technique not a feasible method. From about 10Hz thru 100Hz the coupling caps are 1000uF and the phase shift is only just acceptable, less than 10Hz the phase shift is not acceptable. I'll think of an alternative solution, let you know the outcome. EDIT: Can you say what is the application is for this phase shifted signal, ie: whats the end result.? __________________ Eric " Good enough is Perfect " I will NOT answer PM's requesting technical help, please use the Forum PIC tutorials: Nigel's www.winpicprog.co.uk/ Bill's: www.blueroomelectronics.com/ Last edited by ericgibbs; 7th July 2009 at 01:09 PM.

7th July 2009, 01:26 PM	#20
meastwood98	We have a linear sensor that gives out two sinusoidal signals 90 degrees apart, these are put through an interpolator (which we cannot play with the electronics inside) and then into a PC for Transmission Error Measurement. The problem we are facing is a large error on the PC, which is caused by the two signals not being exactly 90 degrees apart. There is approximately 2-3 degrees of error.

8th July 2009, 07:37 AM	#23
meastwood98	The problem is that the interpolator and the computer software is something that is a permanent setup that we use with various different linear and rotarty encoders. Therefore the solution we want is to be able to just plug the sensors in as we would with an encoder without editing any software or the interpolator. The sensors themselves only cost approx £5 and therefore cannot really be expected to give the same accuracy as an expensive encoder, however we were just interested if we could manipulate the signals to get somewhere close. If it cant be done its not a problem, thanks anyway.