[hobomachinist]

I am very attracted to the Acorn system for ease of use and setup. Was talking to a guy at work that sets up and configures our robots and ..well hes a servo guy...hes been helping me plan a cnc milling machine..atleast the control aspect and motors. Told him that i was considering Acorn CNC. ..showed him the first "features" page and he was like ..well its not closed loop. He gave me a crash course of what that means. Is that a big deal? he seemed to think it wasnt ...uhh Ideal? Am i Missing something? I guess i should mention i havent decided on motor selection yet either.

[pippin88]

Most modern servo systems close the loop within the servo drive.

Most servo drives accept step and direction input.

'the loop' being positional (rotational) feedback from the motor and the power output to the motor.

Steppers are (usually) open loop. Power is sent to the motors with no feedback. The drive / controller assumes that the desired movement has occured.
Closed loop systems detect when the desired movement has / has not occurred and correct.

[hobomachinist]

[QUOTE=pippin88;2567284]Most modern servo systems close the loop within the servo drive.

So if I where to select a motor package that had feed back closed loop situation in the drive itself I could still get the position confirmation guarantee.. even though acorns control is open loop? Is that correct? Or am I still missing something?

[gabedrummin]

Acorn is a motion controller .It will send the step and direction signals to the driver . In a closed loop system the controller will correct the step train to make sure a step is not lost . Even if you use closed loop driver acorn will not correct .When a step is missed your motors will stop turning . Acorn is A hobby grade motion controller . A full closed loop system will be thousands of dollars more . Based on the the amount of money you want to spend .Directs you to the system you use.

[joeavaerage]

Hi,
your friend is correct, the Acorn is not closed loop, however the Oak,also from Centroid is.

Wind back thirty or forty years, all controllers were closed loop. This refers to the encoder (or linear scale or resolver) feedback be applied to the controller, and the controller making signal
alterations called PID (Proportional, Integral and Differential) which would be amplified and applied to the servo. Such systems were, and still are, expensive, complex and can be very hard to tune.
But that was all that was available.

From say the 90's servos became available with drives that were capable of closing the loop. With such a servo you could issue a movement command and the servo and drive would enact that command.
The drive has feedback from the servo mounted encoder and it drives the servo using the same PID control scheme as the previous generation servos. The PID loop gave such servos every bit as good
performance as the earlier system, and in many cases better. The manufacturer of the drive also makes the servo and therefore has extremely detailed and precise understanding of the PID control necessary to get the
very best from the servo. A generic 'closed loop controller', here I'm referring to the earlier generation type, could not hope to control all and any given servos as well.

With the advent of these closed loop servo/drive simple Step/Direction controllers were now not only possible but very competitive with the older style closed loop controller. Amongst the solutions used by
hobbyists you will no doubt have heard of Mach (3 or 4), UCCNC, PlanetCNC, Centroid Acorn, Edging.....and this is just a small sampling of the better known solutions, and they are all Step/Direction controls,
that is to say NOT closed loop controls. These solutions are much simpler and cheaper than the traditional closed loop controllers and have come to dominate, and they rely on the servo and drive being
closed loop rather than the controller.

There are still closed loop controllers out there. Probably the best of the lot is and most widely known is Galil. A three axis Gaili controller (current model) will set you back something like $2500USD....so not cheap.
There are many others, less well known, and probably lesser performing but none the less good devices like Masso, Centroid Oak, Dynamotion KFlop and others.

You might wonder if closed loop controllers are so expensive and do not provide any performance advantage over Step/Direction controllers with closed loop servos why would they still exist?
Virtually all servos require a rotary encoder, whereas many industrial CNCs have linear scales, and then closed loop controllers come into there own. They are still very useful if you are trying
to re-use an old machine. Many and old mill from the 80's is still a classy and solid machine, and if you can resurrect the old control system still make for a great machine, old or not.

As it turns out there are specialist servos and drives that can use linear scales as well, but for the moment lets not add any more confusion.

In a closed loop system the controller will correct the step train to make sure a step is not lost . Even if you use closed loop driver acorn will not correct .When a step is missed your motors will stop turning .

This I think is a little misleading. It is certainly correct the a Step/Direction controller like Acorn cannot take any corrective actions should servo lag....but this is the whole point of feedback servos, its the drive that takes
the corrective action. If it turns out the the drive cannot catch up, because of an overload or jam or something, then it will recognise that its slipped too far and fault out and signal a fault to the main controller
which will then stop the whole machine. Even if you had a closed loop control the ability of the servo to catch up is still limited by the same overload or jam. Should a close loop controller encounter an axis servo
that cannot keep up (overload or whatever) it will recognise the lagging servo and stop the whole machine.

In this regard I don't think the a Step/Direction controller (with feedback loop servos and drives) are any worse than a closed loop controller, and that is what gabedrummin's post seems to imply.
Apologies to gabedrummin if you think my criticism is to nit-picking.

I have used 750W Delta B2 series servos on my mill. They are closed loop, that is to say the servo encoder is feedback to the drive, and the drive ensures any corrective action require to faithfully
enact the Step/Direction commands fed to the drive from my Mach4 controller. My Mach4 controller is very similar in style and performance to an Acorn. In the two years I've been running this
machine I've never had a 'following error fault' Following error is when if a servo lags too far behind the commanded position it faults out. I've had a few stuff-ups and crashes....and then I might
get a servo fault....but then I'd get the same sort of fault with a closed loop controller as well, I mean that if I stuff-up and allow a tool path which collides with a clamp, it does not matter whether its
closed loop or Step/Direction its still going to crash!

https://www.fasttobuy.com/flange-80m...er_p28084.html

My advice to OP is that a closed loop controller like Centroid Oak does not offer any advantages that justify the extra cost over a Centroid Acorn. If you were to use the difference in cost to fund closed loop servos
(I'm not a fan of closed loop steppers....for various reasons) I think that is a better use of your resources.

Craig

[hobomachinist]

Hello Craig, it's good to cross paths again. Your input was appreciated and provided the clarity I sought. In a previous post, I mentioned receiving three Moog servos from my colleagues at work. While they were a generous gesture, I found myself not drawn to them. Time is valuable, and I decided against investing it in understanding these servos, even if they came at no cost. You rightly pointed out concerns regarding documentation and the potential difficulty and cost of obtaining compatible drives. However, this isn't insurmountable, given that I have access to such drives at my workplace.

The individuals assisting me at work - let me rephrase that, discussing my project and offering spare parts from machines that get upgraded controls and motors - believe that assembling these leftovers is an obvious cost-saving measure..Yet, I find myself aligned with your perspective. I prefer commencing with a control system that has a community, comprehensive documentation, reasonable pricing, and similar considerations for the servos, including warranties even.

I was seeking reassurance that I could achieve the feedback and location precision I desire.

Also, I wasn't aware that there were two types of closed-loop systems: controls and/vs drives (motors). I must admit, I was initially enthralled by the idea of Acorn control, only to be somewhat disheartened when my colleagues dismissed it as non-closed loop. However, they remain under the impression that I'll be using the older Moog servos. I find it difficult to express that I never truly desired them, but they were a gift and are now in my possession. Hence, I've resolved to utilize them in funding my DIY CNC mill, roughly the size of a Tormach 440.

I believe I've finally settled on the Acorn control system. Now, I just need to identify servos that align with my preferences and come with thorough documentation. I anticipate requiring guidance in the electrical aspects of this undertaking.

Once again, thank you for your insights.

[joeavaerage]

Hi,

only to be somewhat disheartened when my colleagues dismissed it as non-closed loop. However, they remain under the impression that I'll be using the older Moog servos.

It is entirely possible that those servos were matched to transconductance amplifiers and therefore required a closed loop control. That is to say that these servos are of that previous generation
for which a close loop controller is required.

Such a system may be old, but were capable of very good performance in their day, and still perform well by todays standards. If you have freely available hardware it may well be worth considering. I will
point out that the extra complication of 'closing the loop' yourself and tuning is a learning curve, but not insurmountable.

There seems, to me at least, that there are a group of people whom have the idea that a closed loop controller is superior to closed loop servos. I am not of that group. For reasons which I have posted
they are equivalent, without either being superior nor inferior to the other.

Indeed many of the worlds leading machine maker have gone one step further......bus communication protocol and distributed motion control, and that does away with a controller altogether.

The trend has been for servos and their matching drives to become ever more intelligent. It started back when servo drives became closed loop with the servo. This means that the drive was
a lot more 'intelligent' than the transconductance amplifiers that had been used prior. The increase in servo drive intelligence meant that the controller had less to do, or needed be less intelligent.
The next step is a real-time bus communication protocol like Ethercat, Profibus, CanOpen and any number of other proprietary protocols. Under such a system the PC sends out motion commands
to all the various servos (called nodes) and each servo enacts its portion of the overall motion system. This is called distributed motion control and all that intelligence is built into the servo drive....and now you
don't need a central motion controller at all, not closed loop, not Step/Direction......just the PC.

In the case of Ethercat each servo drive or node is daisy chained to the next node, be it another servo or some Ethercat IO node. 100 of such nodes can be daisy chained together. You can imagine that may encompass several
machines with conveyers and other ancillary automation components operating from just the one PC. It is proving to be very popular among the world leaders in machine automation. It can be safely assumed
they are not of the group that believes full closed loop is the only answer......in fact most emphatically not!.

When I was designing and building my current mill I considered very carefully whether I should go Ethercat. For relatively simple machines such as mine Ethercat does not have any stand out advantages, and yet
there is a cost premium associated with it....and really it would be just for bragging rights. I don't think I would have gained more accuracy, speed or anything else. It would have been a good learning experience,
and one day I might persue it. The servos cost about an extra $50-$100 each, and while I could use Mach4 still I would need runtime licenses for Kingstar Ethercat and Interval Zero realtime core. The cost of the extra
licenses worked out about $600USD more than what I had already and I decided aganst it.

Mach4 and my Ethernet SmoothStepper, an open loop Step/Direction controller, matched with decent closed loop AC servos have proven to be all I could want and more, I have yet to scratch the surface of the capability
that I have without sweating about even more advanced stuff.

I will point out that many an armchair critic will say 'you should do that, or buy this fully closed loop controller, or have linear servos' or whatever, but none of them has ever bought or done anything like what they tell you to do. If you encounter
such a person ask them to show you what they have done.......and the answer is usually they have not made a dammed thing! For those whom have done as they are suggesting to you, look very closely at what they have done
and why. They are the true gems, learn everything you can from them.

Crig

[Al_The_Man]

Just to put another spin on it and to come out in defense of the likes of Galil and Acroloop, which are by NO means dead.
Back when I got into retro-fitting CNC M/C's in the '80,s , these were the only DIY game in town for Cards that sat in a PC slot and provided up to 8 axis CNC.
I pretty much stuck with Galil, after Acroloop was taken over by Parker-Hannifin.
One nice feature of Galil, is the feature of electronic gearing and electronic cam, of up to 8 axis off of one master encoder.
Another, was the ability to use relatively cheap simpler analogue (transconductance) drives due to the controller closing the loops, at that time the feedback rate was ~12mhz, now they have the ability to close that loop at 22Mhz.
You can also still pick up the odd Galil card on Ebay etc, that will still more than do the job of any DIY CNC M/C,.
All my customers were industrial manufacturers so money was not that great an issue as I could give them a DIY system much cheaper than the vendors they were used to.!
.

[samco]

Raises hand...

I can only give you educated opinions on closed loop servo/stepper drive combos.. (I have never actually used any of said drives - so you see where I am coming from)

If they are configured correctly and are quality designed - closed loop drives should work perfectly fine... If the control has a good Trajectory planner (I have never heard anything bad about Centorid - they are a big player in the industrial control market) then everything should work just fine..

that being said.. if the drive doesn't follow the trajectory of the controller - will you know? (then probably does it matter or do you care - but that is a bigger question)

Can you set the following error in the drive so it trips an estop when the following error is over a certain amount?
Look at something like a gecko servo drive(g320x). Its minimum following error is +/- 256 encoder counts. This is how far out the drive position needs to be before it will error.. That seems like a lot..
Here is another https://www.cnczone.com/forums/dmm-t...-handling.html

I really wonder how well some setups are actually following the commanded trajectory.

I guess what I am saying - acorn -> step/dir -> closed loop servo/stepper drive if configured correctly with good drives that actually follow the given trajectory well - should work just fine.

This is no difference from a good stepper setup. it will follow the command trajectory just fine until it stalls. A decently configured open loop stepper system works just fine too...

(I am partial to closing the loop in the control though - using linuxcnc - but that is another learning curve.)

sam

[joeavaerage]

Hi Al-the-Man,

Just to put another spin on it and to come out in defense of the likes of Galil and Acroloop, which are by NO means dead.

I certainly did not claim that closed loop controllers were dead, but it is certainly true that in the 80's that was all there was but now closed loop controllers make up
small fraction of the market. It's probably argumentative to say they have been superseded but that is in fact what has happened. The areas where closed loop controllers
remain exploit some of the features you have highlighted and/or as repair/refurbishment of older machines, both are valuable sectors of the market, but much diminished from yesteryear.

I have looked keenly at Galil over the years, and while I like the quality and flexibility the price is too high. I am very reluctant to spend moderate sums, say $1000 or so on a secondhand Galil, when more often than not
the vendor cannot demonstrate that it's still working. If I were buying Galil I would buy new or not at all....and that's the rub, a new 5axis Galil (DMC4050) is $2990USD

By comparison I could buy a refurbished PC with high-end Ethernet card with a Mach4Hobby license, a Kingstar Mach4Ethercat runtime license and an IntervalZero realtime scheduler runtime license for $1600USD.
As I posted earlier I declined to spend that extra, but I did consider it closely.

What I do have is a cheap PC ($400) with a Mach4Hobby license ($200) and Ethernet SmoothStepper ($225 at current pricing) and a home brew breakout board estimated value equal to that of an MB3 ($190)
for a total of $1015USD.

My existing control system (Mach4/ESS/BoB) is open loop and yet exceeds all my expectations and I have yet to discover any process or procedure that I cannot acomplish with it. Bang for your buck I'm happy.
I do not claim that my system is best or better than full closed loop, nor better than Ethercat, but I claim that I have a perfectly workable system at moderate cost.

Craig

[joeavaerage]

Hi Samco,

that being said.. if the drive doesn't follow the trajectory of the controller - will you know? (then probably does it matter or do you care - but that is a bigger question)

Can you set the following error in the drive so it trips an estop when the following error is over a certain amount?
Look at something like a gecko servo drive(g320x). Its minimum following error is +/- 256 encoder counts. This is how far out the drive position needs to be before it will error.. That seems like a lot..

Yes, the controller can and will know if the drive faults 'Following Error'. Gecko 320's are pretty old school and very basic. That they have a fixed and fairly wide following error window should be no surprise.
Moden AC servo drives, and that includes the cheap as chips Chinese made units flooding the market all have a programmable following error window.

It is common practice to set the window wide at the factory so that a user may tune the servo in the machine without pesky faults. Once the tuning is complete however then you tighten up on the following error
window to suit the tolerance you need to achieve. It is if you like the final part of the tuning process. Essentially if you have a wide window then the expectation is that the servo will easily follow its commanded path
within that wide tolerance and you would not see any following errors except in exteremis. Alternately you may tighten up on the window but then you might expect the servo to occasionally lag to the extent that it
would fault. You might want to reduce the max acceleration in the CNCsoftware a little for instance and give the servo just that little extra time to accurately follow the commanded path.

The ability of a servo to follow a given trajectory is a fundamental property of the physics of the servo and its drive. For instance its torque vs its rotational inertia and whether the drive has the current capacity to achieve
the acceleration the trajectory commands or the voltage withstand to decelerate as the trajectory demands.

Does it matter whether you use a closed loop controller like Galil or LinuxCNC OR rely on the feedback mechanism built into a modern AC servo drive.......the absolute best a servo can do ai related to it
torque, inertia, current and voltage?. My argument is that they are all equivalent assuming they can all maintain the same closed loop bandwidth.

Just as an example I have an older style, still AC servo, by Allen Bradley circa 2005. Amongst the things that you can program into the drive there is a piecewise linear approximation of the current to magnetic saturation curve
of the servo. This allows the FOC algorithm to apply a current correction as a feedforward term, and thereby reduce the feedback required with a consequent increase in overall bandwidth. Have you ever heard of
a closed loop controller that has a nonlinear saturation curve programmed into it? I certainly have not. The point being that because Allen Bradley made both the drive and the servo they are able to match
the two to perfection, better even than Galill or LinuxCNC.

Were I to program LinuxCNC or a Galil I could not hope to exceed the 'smarts' built into the Allen Bradley servo drive....unless I too were to program in the non-linear magnetic saturation curve. Note that
Allen Bradley supply the magnetic saturation data for their servo, it comes with the setup and tuning software. You declare to the drive, in fact you don't even have to do that, the setup software detects the model
servo to which it is attached and automatically supplies the saturation data.

This drive was designed in the late 90's and early 2000's, making it 25 years old. In the intervening time AC servos have only ever gotten better and better.

I studied Electrical Engineering at University in the early 80's. Control Engineering was, and still is a very major part of a degree. At that time servos were still old school, resolvers,
synchro's, transconductance amplifiers. We spent hour upon hour, day upon day experimenting with these devices while predicting and analyzing them with the mathematical theory that
underpins them. I will confess to be rather snobbish that any mere feedback drive could hope to attain the 'excellence' that I had so intimately observed while an undergraduate.
Then to rather prove the point I bought an AC servo (400w Delta) just to test it out and prove that my snobbish position was correct. I was not!. Quite frankly a modern AC servo will outperform
your expectations.

It's not that the old school servos were/are bad, in fact if they are tuned to excellence they are just as good as a modern servo, but a modern servo achieves that state of excellence so much easier.

I'd recommend to you that you try them out too. I do not recommend the cheap as chips Chinese servos, not because they don't work, but because the documentation is fair at best and bad at worst and
more often than no setup software. Delta (Taiwanese brand made in China) or DMM (Canadian brand made in China) on the other hand are good quality, documentation, support and most importantly
setup and tuning software at fair prices. I promise you'll spend hours fiddling wiuth the oscilloscope function...it's fascinating. What we would spend many tedious hours (years ago) can now be done in tens of minutes.

This is no difference from a good stepper setup. it will follow the command trajectory just fine until it stalls. A decently configured open loop stepper system works just fine too...

100% correct, a well designed open loop stepper system is every bit as good PROVIDED the stepper is used within it's capability. Feedback servos may allow you to push that somewhat,
but they too have a limit beyond which they are overmastered. A full closed loop control cannot make that range of operation any better, full closed loop control is good, but it's not magic.

Craig

[suntravel]

Full closed loop with Acorn6.

I am happy with the results.

https://youtu.be/UvlV7wJggpY

Uwe

[Al_The_Man]

I agree there is not really any discernable difference between the two systems of feedback.
The term "closed Loop" applies to all controllers that use a feedback correction,, whether it is done by the controller or the drive, there is not really big advantage to either, although there is a couple in favour of the controller where as I previously mentioned, the drives can be Much simpler, and also that fact that the other features of multi-axis electronic gearing.
On the reference to Allen-Bradley, in my personal experience, they never really took off as a CNC control, in all the years in the business I only recall one such system using AB.
Apart from the DIY Galil for the low cost systems, The upper end system I used in all my retro-fits were Mitsubishi, apart from nice equipment, the company was excellent to work with.
Especially when compared to the likes of Fanuc!!

[phomann]

As I understand it, having closed loop back to the controller, accounts for backlash. Closed loop back to the motor driver does not.
That said, if you are going to the effort to close the loop to the controller, you should have already removed backlash through mechanical means.
Cheers
Peter


Sent from my iPhone using Tapatalk Pro

[pippin88]

As I understand it, having closed loop back to the controller, accounts for backlash. Closed loop back to the motor driver does not.
That said, if you are going to the effort to close the loop to the controller, you should have already removed backlash through mechanical means.
Cheers
Peter


Sent from my iPhone using Tapatalk Pro

Not correct Peter

Backlash depends on the loop

Most closed loop involves an encoder rigidly coupled to the motor. This cannot account for backlash, which occurs in screw/nut, or in couplers between motor and screw.
The encoder feedback can go to the controller (loop closed in controller) or to the drive (loop closed in drive). AC servos / brushless servos need rotary feedback for commutation

It is possible, with certain motors (e.g. brushed DC), to have single closed look with a linear encoder (no rotary encoder). Not done often as response times often poor.

You can have dual closed loop with a rotary encoder on the screw, and a linear encoder. This can deal with backlash as the linear encoder measures the true position of the axis. Sometimes the second loop is back to the controller, sometimes to an advanced servo drive.

I have some Delta A3 servo which have inputs for a linear encoder and are dual closed loop

The first strategy for backlash should always be mechanical - minimise the problem

[joeavaerage]

Hi Peter,

As I understand it, having closed loop back to the controller, accounts for backlash. Closed loop back to the motor driver does not.

Not quite. If you have a linear scale that is called 'load sensing'. Any system which closes the position loop using load sensing will accommodate (within the acceleration/bandwidth limits) backlash.
In the case of a motion controller it seems obvious. Less so with load sensing servos.

All AC servos must have a rotary encoder and it must be direct coupled to the rotation of the armature, any lash in that coupling will severely degrade the bandwidth of the Field Oriented Control (FOC)
algorithm to the extent that the servo will probably fail. Load sensing AC servos have two encoder inputs, the regular rotary input as required by FOC, but it has a secondary encoder, a linear scale for instance,
and that is attached to the load. The position loop is closed by the linear scale and just like a 'full closed loop controller' accommodates backlash. The rotary encoder is required for the torque and velocity loops,
but the linear scale closes the position loop.

Assuming both a 'full closed loop controller' is well designed and executed, and a load sensing AC servos is well designed and executed and both have the same closed loop bandwidth the results are identical.
Subtle variations in design and execution can result in performance differences. The same PID code applies to both identically, and has been industry standard for some decades now.The only differences are
the execution of the differentiator filter bandwidth, the implementation of anti-integrator wind and how limits are applied (soft limits vs hard limits).

This video does an excellent job of explaining how a PID controller is done in software. The surprisng thing is that is is in fact so simple.

https://www.youtube.com/watch?v=zOByx3Izf5U&t=6s

Manufacturers of both AC servos and full closed loop controllers have a distinct interest in making their product perform as best they possibly can and consequently there is very little to pick and choose
between them. If they have the same closed loop bandwidth you cannot tell which is which.

One area where AC servos have a distinct advantage is notch filters. Full closed loop controllers can also have notch filters, but they seldom do, and even more rarely have more than one. AC servos usually have
two self-tuning notch filters.

Craig

[phomann]

Sorry, I was assuming that closed loop to a motor controller has the encoder on the motor. Closed loop to the controller uses an encoder on the axis.

Cheers
Peter


Sent from my iPhone using Tapatalk Pro

[Al_The_Man]

Also: The early DC brushed drives had a dual voltage loop which required first tuning the motor feedback loop via a DC Tach to the drive, and then the PID loop was tuned via the encoder back to the controller. Bit of a pain!
Then came the torque mode (transconductance) current amplifier which eliminated the need for the tach.
I recall when the first of the AC BLDC servo's came out, it was suggested to me by one salesman that you could not use them without gearing for low RPM's due to the 'cogging' effect caused by the very low (commutation) pole count.
What he did not account for was the very high PID rate of feedback, at that time Galil encoder feedback feedback was 12Mhz, now of course, much higher, the result was, BLDC could be made to move at very low RPM's as smooth as any DC motor.

[joeavaerage]

Hi Al,
while a separate tacho generator may be a thing of the past, the same function is still required.

All motors, whether they be brushed DC, BLDC or AC servos are all current driven torque machines. To make a servo the torque machine needs be enclosed in a velocity loop, and to make
it position controlled the velocity loop is itself enclosed by a position loop.

As you point out the tacho generators of yesteryear provided the velocity feedback, and it was that feedback that closed the velocity loop.

In more recent years we have high resolution, high reliability encoders. You can derive the velocity from an encoder by differentiating the position, thus the encoder provides position feedback for the position loop and
also provides velocity feedback for the velocity loop.

The earliest of encoders were low resolution, and presented a real problem for use as a 'velocity measurement' device. Imagine a motor turning very slowly. At a give time the encoder count is nnnn.
At the next sampling time, say 100us later the encoder count is nnnn+1, and so the velocity is (nnnn+1) - nnnn =1/100us. If the motor is turning very slowly then maybe in the next sample time the encoder has not increased at
all, and the velocity would be 0/100us, and the next sample the encoder may have increased by 1 and so the velocity is 1/100us....and so on. The true and actual velocity is about 1/2 a count per sample period, but 1/2 cannot be
represented by a low resolution system. This is called quantization noise, and represents a distinct limit to the ability of an encoder to be used to derive a velocity signal.

As the resolution of encoders got better and better and the sample time (equivalently the processing power of the micro controller) decreased this limit increased to the extent it was as good or better than the bandwidth of a tacho generator....
and therefore tacho generators fell out of favour.

If you have followed the video I posted you will see that the differentiator is followed by a single pole low pass filter. Over the years many have tried multipole filters but have run into phase margin faults, ie instability, others
have tried equi-phase two pole filters to avoid the phase margin degradation....and others have tried every manner of filtering regime, with and without lead/lag compensation, with none really providing any better performance
than the simple single pole of the video.

By far and the most common way to derive a velocity signal from an encoder is to use an integrator in a feedback loop....which has the effect of a differentiator in the forward path, but results in 6dB better noise performance.

So its not that tacho generators are gone, but rather that encoders are used to fill the role that tacho generators filled.

Craig

[Al_The_Man]

Hi Al,

So its not that tacho generators are gone, but rather that encoders are used to fill the role that tacho generators filled.

Craig

All the DC voltage loop drives I tuned, also had an encoder, generally a min. of ~2k pulses/rev, (then using the quadrature count x4 ). these were used in the vast majority of early systems. as I understood it, the transconductance/torque mode amp rather than voltage loop replaced them.