Control.com - Nerds in Control

I design a interface circuit of sin/cos encoder,
but i don't know the amplitude of its output. Do different companies produce this kind of encoder with the same output voltage amplitude?

> I design a interface circuit of sin/cos encoder ... <

You say sin/cos....and encoder???

encoders have a digital output and are either on or off. they come in quadrature, single channel, dual channel with open collector outputs or ttl outputs. linear or rotery. also they can be incremental or absolute. there are many more options to choose from including, package shaft size, pulses per revolution, ect. a catalog from an encoder manufacturer will list and explain some of these variables.

resolvers use a phase difference between the reference, sin, and cosin waveforms. the lead, lag,and frequency can be decoded into forward, reverse, and velocity information by comparing the sin & cos output phasing to the reference phasing.
resolvers can be rotary, or linear, and have many packages and shaft sizes, and poles.

to answer your question, id depends on the specifications of the particular encoder you are using.

hope this helps
Tim

I haven't looked at this particular encoder, but most of the sin/cos encoders have 1 volt peak to peak. This is extremely susceptible to noise, and will require some care in wiring, but works great when used properly. I'm sure that Heidenhain has good recommendations for wiring. Basically, use double shielded twisted pair cables, and don't run them in the same trays as your motor cables. Also, watch your grounding scheme very carefully.

Davis Gentry
Delta Tau Data Systems

Heidenhain optical encoders use
two different amplitudes
1) 11 microAmpere peak to peak
(NOT milli amperes, MICRO ampere)
and that's for the A&B phases,
it's only 5 uAmp for the Z (home) phase
2) 1 volt peak to peak ( not so common in my field)
i think these are for noisier environments or
longer distance.
This may be the new 'standard'
for machinery use (what Heid reccommends)

btw: Heid SELLS a pc card to handle 2 or 4 axis
(i think, this info is a couple years old)
IF your heidenhain local agent suggests buying an
'EXE' box (xlates analog sinusoidal input to
digital and multiplies it too) RUN AWAY, HANG UP,
way too expensive, unless you have lots o money.

> I design a interface circuit of sin/cos encoder, but i don't know the amplitude of its output. Do different companies produce this kind of encoder with the same output voltage amplitude?<

Sin/Cos encoders produce about 1v ptp as viewed on an osciloscope - about .31 rms.

The interpolator box is a analog to digital converter with a mulitiplication feature. A 10X box gives ten times the number of counts per rev. The boxes are very expensive, $600+ per SIN/COS encoder, but allow for very high resolution. The multiplier factor is fixed so that at higher speeds, you will saturate your controller.

Another approach is to use the ACS-Tech80 SPii+ card which multiplies in hardware on the card. You can adjust the multiplier on-the fly through software. You can make a high-speed move with low multiplication and then up the multiplier to capture that sub-nanometer position accuracy.

Bob
Apps Engineer
www.acs-tech80.com

Heidenhain lists their ERN 1387 encoder as having 1Vpp output. Some of their older sin/cos encoders use the 11uA format, but they consider this obsolete. All new designs sport the 1Vpp output for sin/cos.

Heidenhain's leadership and clout in this field means that this is the de facto standard. Not everyone follows it, but increasingly other companies are falling into line (even as their designers grumble that the signal should be bigger).

Our older interpolator circuits had a bank of jumpers to accommodate a bunch of different levels; our newer ones take only 1Vpp.

Curt Wilson
Delta Tau Data Systems

P.S. Tim: The digital quadrature encoders that most people use are simply analog encoders with comparators on the outputs to square them up. This makes for a simple digital interface, but limits you to 4 states per line. As people want better resolution, it turns out that it is more cost effective to get additional states per line than it is to get additional lines per rev or per mm. To do this, you need to get at the analog signals directly and process them in a more sophisticated fashion.

We manufacture a sin/cos encoder that provides a differential 1Vpp signal. This seems to be somewhat standard.

The choice to start manufacturing sin/cos came after several requests from customers to get direct access to the raw analog signals our incremental encoders directly from the sensor. The normal variances in signal amplitudes over speed and temperature made using signals directly from a sensor less desirable, so we opted to develop a sin/cos encoder that outputs robust 1vPP signal to interpolate from.

Feel free to shoot me an e-mail if you have any questions.

You can check out our spec sheet for the SC12 sin/cos encoder we manufacture at:
http://www.quantumdev.com/products/optical_encoders/sc12.html
< br>Jim Miller
Application Engineer
Quantum Devices Inc.
jmiller at quantumdev. com

I looked at the data sheet. The output line driver for the sine/cos signals is listed as 26C31 type, which is an RS-422 digital driver. I think perhaps this is a misprint.

The 26C31 is for the Commutation channels.

Dear JIM,

I'm looking out for Sin/Cos Encoder working principle. Where can I get this? Kindly guide.

Chinmay S. Pemgirikar
Project Engineer
Electronica Machine Tools Ltd,
csp_emtl at electronicagroup. com

Chinmay,

I am not sure what you mean exactly.

It sounds as though you are looking for a description of how a SIN/COS encoder functions and perhaps what the advantages might be?

The sin/cos encoder is very similar to an incremental encoder with the main difference being the quadrature signals are sinusoidal as opposed to square waves.

As the previous poster indicated, the advantage of a sinusoidal signal is that the end user (or Drive) can interpolate off of the analog signal the degree of resolution that they need.

If you are asking about the differences in the structure of the optical bench (light, disk and sensor set up) between SIN/COS and incremental encoders, I can tell you that there is very little. The raw signals off of the patented interlaced sensor we developed are sinusoidal, so instead of squaring them, we reinforce them for connection to the outside world.

The SIN/COS index signal may or may not be sinusoidal. Often times this is an option.

If you are looking to develop your own encoder. Besides manufacturing and selling encoders, we do provide sensors for other encoder companies.

I am not sure if this is the type of information you were looking for. If not please clarify.

Kind Regards,

Jim Miller
Application Engineer
Quantum Devices Inc.
www.quantumdev.com

Main reason for using of analog sin/cos 1V p-p is significantly increasing of encoder resolution. As a rule, an interpolation is performed inside (!) motion controllers up to 8192 factor.

Such option allows to transfer encoder output signals with lower (vs. quadrature A,B) frequency and reach higher final resolution. Few motion controller manufacturers only have such option - ACS, Mega-F, Galil...