Cheap o-scope recommendations?

[DICKEYBIRD]

I want to test the encoder on my ORAC lathe with Centroid Acorn control board. Here is the spec: "Acorn accepts industry standard RS422 type differential quadrature encoder output with A, B, and Z channels (Note: low encoder count creates poor performance and accuracy. Centroid recommends 8,000 to 40,000 encoder counts per revolution (2000 to 10,000 “Line”) for most applications."

Is there a $50-ish scope that will do the job?

[Sparky_NY]

You can do a quick test with a digital voltmeter. When A is positive, B is at ground and vise versa. Just turn the encoder a bit and monitor the outputs on the voltmeter. The Z gives a pulse once per revolution, it will be a very short duration and hard to measure on the meter BUT you will probably see the meter reading "blip" as the pulse goes by. It will be very hard to get the encoder to stop so a given output is positive when you want but the key is that when one is positive the other is not. Although a scope is preferred, the above method works well for a quick functionality check. Encoder outputs are typically 5V but thats not a hard rule.

As for a scope, there are some inexpensive adapters that allow you to use a computer as a scope. Here is a example

Seach ebay for item # 330731251939 Its $22

[DICKEYBIRD]

Thanks for the tips George. I know it's functioning because the RPM is displaying. It fluctuates by (what I think is) a normal amount but I'd like to see the actual signal that the Acorn input is looking at.

Before you & Sam jump on the LinuxCNC/HAL scope thing, I promise to go there when I return in my next incarnation. Too old for it now!

[Sparky_NY]

Thanks for the tips George. I know it's functioning because the RPM is displaying. It fluctuates by (what I think is) a normal amount but I'd like to see the actual signal that the Acorn input is looking at.

Before you & Sam jump on the LinuxCNC/HAL scope thing, I promise to go there when I return in my next incarnation. Too old for it now!

Thats funny ! How did you know I was tempted to mention Linuxcnc's halscope?

As long as the encoder is working ok, there is not much to see or check with a scope. You didn't say WHY you want to look at the encoder outputs, normally its only for troubleshooting purposes.

A point of interest, which you probably already realized, is that IF there is a encoder problem of any sort the most controllers will throw up a error message. I went through this with my Dynapath control on my cnc lathe last year when a encoder went bad. The A and B outputs were ok but it lost the Z output (once per rev). The DRO's read properly but the axis threw a error when homing (only time the Z is used on that lathe/control).

Does the Acorn have troubleshooting screens where you can look at the state of the various inputs? My Dynapath had that feature and made troubleshooting easy. Yes, linuxcnc has that feature also.

Although I love linuxcnc, the Acorn is pretty impressive, especially for someone that does not want to get too deep technically. Its quite a nice solution. I have a electronics background and am a geek at heart thus my love of linuxcnc.

[Sparky_NY]

A point of interest.... some encoders have TTL level outputs, others have RS422 line driver outputs. That causes me some grief a few years back. The line driver outputs are slightly lower voltage, with a +- spec like everything. The line driver output was right at the threshold the I/O board I was using would trigger at, that board expecting TTL level inputs. A encoder with line driver outputs may or may not trigger a TTL input its a matter of a tenth or two of a volt and the tolerance of the encoder. I am on to that one now ! Just figured I would pass it along in case you shop for any encoders, watch what type of output they have, same with what type of input the control board expects !

[DICKEYBIRD]

I've wanted a 'scope ever since the 50's when I "helped" my father build a Heathkit that he used in his electronics side jobs. You'd think I'd be a chip off the old block electronics whiz but alas, the car bug bit me & that was that.

I checked Amazon first & I saw a bunch of 'em but I don't know what bandwith/sample rate I need for the encoder testing & also simple auto tests.

[MaxHeadRoom]

I got me a Syscomp double 'beam' electronic unit some years ago when they first came out, USB plug in unit.
For the majority of work in audio, CNC or general electronics, they are great, Also you have the ability to store the display.
With a lap-top, makes it light-weight portable.
Max.

[Sparky_NY]

I've wanted a 'scope ever since the 50's when I "helped" my father build a Heathkit that he used in his electronics side jobs. You'd think I'd be a chip off the old block electronics whiz but alas, the car bug bit me & that was that.

I checked Amazon first & I saw a bunch of 'em but I don't know what bandwith/sample rate I need for the encoder testing & also simple auto tests.

With bandwidth more is better but usually comes with increased cost. For your intents anything 10mhz or up should do the trick. No doubt there are youtube reviews of some of the computer scope adapters, that would not be a bad place to start looking.

Be warned, there is a learning curve to understanding use of a scope ! Understanding what you are seeing is the trick as well as configuring it to catch what you want to see. I would say to get a inexpensive one and get your feet wet, its not a big investment and you can upgrade later if needed.

[MaxHeadRoom]

This is the one I have, although the Mark1 ver.

https://www.syscompdesign.com/product/cgm-101
Max.

[PStechPaul]

There are several here that are $50 and up. Many of them are also DMMs.

https://www.banggood.com/search/mult....html?from=nav

Here are some (mostly) in kit form for as little as $13:

https://www.banggood.com/search/osci...e-diy-kit.html

You can get used "old school" scopes on eBay for around $100. Here is a Tek 100 MHz storage scope for $120 plus $20 shipping:

https://www.ebay.com/itm/Tektronix-4...e/324330748569

[RB211]

You could also buy a rotary encoder that is known to work well. Centroid sells them.

[andywander]

Y...When A is positive, B is at ground and vise versa....

In what universe?

Seriously, encoder quadrature outputs are supposed to be 90 degrees out of phase, so while the transitions of each output will never occur at the same time, there will be times when both A and B are positive, and times when they are both at ground.

Or am I misunderstanding something here?

[Paul Alciatore]

No, you are not misunderstanding anything. The idea of two outputs that are in a quadrature relationship with each other is how an encoder provides information not only on position and motion, but also on the direction of that motion.

The question of what scope you need to check these outputs depends a lot on what characteristic of them you want to check. If you just want to check the levels, then you can just use a VOM and check the levels with the encoder at rest (not turning or moving). That would reveal some problems, but not all.

If you want to check the quadrature nature of the signals, you could do so with two VOMs or even two lamps (LEDs). Just turn/move the encoder very slowly and observe the needles jump up and down or the lights come on and off.

So far, a scope is a luxury, not a necessity. It is hardly worth mentioning in today's world, but for the observations above a TWO channel scope is what is needed: one channel for each of the encoder's quadrature outputs. Virtually every scope I have ever seen has had two channels: that goes all the way back to the Heathkit scope that I built about 50 years ago. Every digital scope that I have seen advertised today also has at least two channels. Note: two channels means that you can connect two probes to the scope at once and see both signals on the display at the same time. This is the way you see the time relationship of them and hence observe the quadrature nature of an encoder's outputs.

Now, there are other things that can affect the encoder's performance. Things like what happens when there is movement/rotation of the encoder? What happens when that movement/rotation is fast? Real fast? There are fancy, digital terms for the problems that can occur, but back in the audio days it was called frequency response. And scopes do have frequency limitations.

If the encoder or the connections from it to the controller do not have sufficient frequency response then the digital signals will not be a proper square wave. The edges of the square wave start to become rounded and the square wave starts to approach a sinusoidal form. In extreme cases, not only does the square wave become a sine wave, but the amplitude of that sine wave will start to drop off. If that happens, then at some point, it will no longer be in spec. for the controller's inputs and counts will be lost. This would happen when the encoder is moved or rotated fast.

If you want to see this last type of problem, which is related to frequency response, then you need a scope that can display signals of higher frequency. If you really want to see the shape of the waveforms from the encoder and not have the scope itself round them, then the scope will need to have a frequency range that goes as much as TEN TIMES as high as the frequency of the square waves from the encoder. Looking at the 8,000 to 40,000 encoder pulses per revolution you mention in your post and if the encoder only makes one revolution per second, that brings us to a range from 8,000 to 40,000 Hz. Splitting the difference to 20,000 Hz and multiplying by a factor of ten you find that you need a scope that has a response of 200 KHz. This is not an excessively high figure as many inexpensive scopes will go up to 5 or 10 MHz. But remember I used a 1 RPS or 60 RPM number to derive this. If your encoder goes faster than this, then it must be multiplied upwards.

Another consideration is triggering. In scope talk, triggering is when and how the beginning of a trace across the screen begins. The thing that must be kept in mind here if that a scope does not just display one trace and then stop. Scopes, even digital ones, usually continuously display the signal. When the trace reaches the right edge of the screen, it jumps back to the left and starts all over again. This can happen many times each second: when displaying the 200 KHz signal mentioned above, this could be as many as 100,000 traces each second. If this triggering or start time is random, then those 100,000 individual and non-synchronized traces will be "on top" of each other and you will probably have just a big blur. You will not be able to see any details at all.

Most scopes have some kind of triggering which means that the traces will all start at the same point on the waveform. Then the successive traces will be in sync and you can discern the details present in them. There are also single trace modes on some scopes, but these tend to be more expensive and some form of triggering is still used for that single trace. Good triggering is a must and I have had many triggering problems with scopes that cost $2,000, $4,000, $8,000 and even more. If your signal repeats exactly from one cycle to the next then triggering can be easy. But if you are trying to see a problem that occurs only once out of every 10, 50, 100, or 1000 cycles, then triggering will be your friend or your enemy. An example that occurs here would be an optical disk encoder where one of the 8,000 holes in the disk is too small or in the wrong place. That means you would need to trigger the scope so that this bad hole is shown on the screen in the same place each time the disk rotates. You would need a once per revolution trigger which would probably be a single hole circle with a third encoder output which would go to a dedicated third input on your scope for triggering. Many times I have use three channels on a four channel scope to observe two of them on the screen while triggering from the third. Often this also needed what is called a delayed trigger with a magnified trace so that one part of a long interval between triggers can be examined in detail. Many professional scopes accomplish this with a dual time base for the sweep: one sweep rate for the entire time between triggers and the other, faster sweep rate to show the details of the selected part of that time frame. These are capabilities that I have not seen in the inexpensive digital scopes.

One more consideration with digital scopes is the sampling frequency. Much is said about Nyquist–Shannon sampling and many people tend to think of that as the be-all and end-all of digital sampling. This may work for things like audio response and your ear may not hear any more differences if you go beyond that point, but if you want to see problems in a waveform, especially a digital waveform, then you will need to go significantly beyond that point. Exactly how much further is not easy to say, but again the 10 times number comes to mind and for some purposes, even that may be far short. Going back to my example above, for a real 200 KHz response on a digital scope, 2,000,000 samples per second is the least that I would accept. And remember, that was only for an encoder that was rotating at 1 RPS (60 RPM).

In what universe?

Seriously, encoder quadrature outputs are supposed to be 90 degrees out of phase, so while the transitions of each output will never occur at the same time, there will be times when both A and B are positive, and times when they are both at ground.

Or am I misunderstanding something here?

[J Tiers]

..............


If you want to see this last type of problem, which is related to frequency response, then you need a scope that can display signals of higher frequency. If you really want to see the shape of the waveforms from the encoder and not have the scope itself round them, then the scope will need to have a frequency range that goes as much as TEN TIMES as high as the frequency of the square waves from the encoder.

....................

That is really not true. It LOOKS true, but actually it is misleading.

The fundamental issue is misunderstanding frequency and frequency response. You alluded to that earlier in the post, but then seem to ignore it in the bit above. For this I am assuming the sampling is sufficiently above the frequency response, which is generally true.

The frequency response you need is NOT related to the pulse frequency. It is the highest frequency in the waveform. A square wave theoretically has an infinite frequency content. But because it is produced by some physical device that itself has limits, the upper frequency is limited. The difference shows up in the "sharpness" of the "corners", and the slope of the rising edge.

To get the correct "picture" of the square wave, you must have a 'scope that has a frequency response of at least that highest frequency. If your 'scope is not of at least that capability, you will see a picture that has blunter corners, rounded off instead of square, and also the rising and falling edges will be sloped instead of almost straight up and down.

Those sorts of distortion can affect needed measurements, producing errors in apparent timing, apparent rise time, etc, which may be specified factors that you need to know.

The "rule of thumb" is to have a frequency response 10x the pulse frequency. But that is just an estimate, something that is "usually true" for "typical measurements".

If you need accurate rise time, or relative timing measurements, you may need considerably wider frequency response than the 10x rule gives. You will need to base the frequency response on the rise times you need to measure, and not on the pulse frequencies.

[Black Forest]

In building a CNC machine what types of things would a scope be useful for to diagnose issues? Noise in the electrical circuits seems to be one of the biggest issues in CNC machines and tricky to isolate. VFD's being a big cause of noise. What scope would be good for this type of work? A scope that would save time in diagnosing issues and not just the most inexpensive scope. I would rather buy a better scope and save time which is my most precious commodity.