[Jay C]

Bruce emailed me something a while back as an idea for those of us in the need for really long scales. I will try to upload it unless it is copywrited, then I will just have to make a diamgram and upload that instead.

Mike

Good news on your wife.

Bruce posted about his concept for a less expensive encoder in this post. Unfortunately, in the email Bruce sent ... I'm the wrong kind of engineer to grasp the details

Jay

[turmite]

Good news on your wife.

Bruce posted about his concept for a less expensive encoder in this post. Unfortunately, in the email Bruce sent ... I'm the wrong kind of engineer to grasp the details

Jay

Hi Jay, nope that's not it.....well it may be, but mine is a picture....so much easier to understand.

Mike

[turmite]

Here is the concept that Bruce made in Rhino. The image is a little rough due to my having to cut it, but hopefully it will make sense once you see it.

Mike

[LUCKY13]

Jess, you might be confusing me and Dougal in NZ or someone else at the very least. I was merely confirming for my own sanity the encoder interface. The "manual" I'm referring to is the software one can download for the Brain. I just ran it in Demo mode. If you click on the IO tab you'll see what I see. I never saw a link for a user manual.

Jay

Here is a user manual, it had to be zipped, hope I did it right.



Yea sure enough the Demo has the typo, hopefully that want cause anyone any problems.


Jess

[Jay C]

Here is the concept that Bruce made in Rhino. The image is a little rough due to my having to cut it, but hopefully it will make sense once you see it.

Mike

Yeah, that wasn't what he described at all. This looks more like a toothed belt driving another toothed belt with drives an encoder. I don't know enough to ask proper questions, but I'm not sure how this is better/different from a rack and pinion?

No, what he described was magnetoresistive with a 5micron field using "old vernier caliper tricks to get the resolution". I feel bad not being able to keep up with conversation he and I were having ... he is on a whole other plane with his thoughts, ideas, and experience. My brain ached when I went back and re-read what he wrote

Jay

[Jay C]

Here is a user manual, it had to be zipped, hope I did it right.

Well, based on the information in the users manual, all signals to/from the Brain need to be opto-isolated. Did you say you were using the C10 BOB? If so, you are good there, but you will still need optos on all your encoders due to the 3.3V operating voltage of the Brain.

Jay

[Dougal]

Well, based on the information in the users manual, all signals to/from the Brain need to be opto-isolated. Did you say you were using the C10 BOB? If so, you are good there, but you will still need optos on all your encoders due to the 3.3V operating voltage of the Brain.

Jay

Does this post help this discussion?
http://www.safeguardrobotics.com/for...W=&PID=138#138

Hopefully you guys will have all this nutted out by the time I'm ready to order some scales.

[Al_The_Man]

I can see the use of opto isolation on general I/O to enable the use of industry standard 24vdc for the I/O as is laid out in the manual.
But in the case of encoders, the mention that the internal 5vdc can be used on the other side of the opto defeats the isolation anyway.
In my opinion it would have been much better from a design consideration to use a differential receiver together with use of the internal 5v supply.
Each Encoder circuitry is virtually self-contained anyway, When differential circuitry is used together with a professional style cable such as Belden 9881, there should be no need whatsover for opto isolation.
This now turns it in to a pain in the neck if for all encoders, a opto interface and differential receiver has to be supplied.
Al.

[LUCKY13]

Well, based on the information in the users manual, all signals to/from the Brain need to be opto-isolated. Did you say you were using the C10 BOB? If so, you are good there, but you will still need optos on all your encoders due to the 3.3V operating voltage of the Brain.

Jay

Jay, I haven't bought any BOB yet. In fact I am no where close to hooking up my Brain ( its not arrived yet anyway). I believe I will kinda wait things out before I do any wiring, I still have enough mechanical stuff to work through on the mill for now so I am in no hurry (the kinda stuff I understand well).


I am still kinda lost on this TTL stuff though. I can understand why they would be needed but after researching how this TTL stuff works I don't understand how Sandefuj got it working with one piece being a 3.3v system and the other supplying 5v signals. I guess the encoder is supplying a ground signal and thats how it still works but it seems like that would still overload it. I hope Sandefuj doesn't have any problems with it later. I found a little Opto board that is very cheap that claims its purpose is for this type app , (3.3v to 5v system compatibility). It also claims to be very high speed with 8 optos and only cost about $5.00. If anyone is interested in this little board here is a link because I am not sure whats the best way to do this (with or without optos). I don't know how good this opto board is either but you guys can look and maybe even let the rest of know if its worthy.


Opto Board





The main thing to me even though it may be a pain is for it to work without letting the smoke out or causing any problems with timing/speed of the encoder signal. After reading up on differential signals I can understand where Al The Man is coming from. I expect that sometimes the environment in my shop will get a little noisy, differential encoders would probably be a best bet. I guess I will look into this a little deeper before I even think about buying position encoders. IMHO I will probably start out with no closed loop so I can simplify things until I am up and running, but closed loop is the goal.



Jess

[sandefuj]

Jess:

If you look at the output of the Brain on pin 7 on both of the Motor and Glass Scale it is 5V just what you want, no 3.3V, so all of you Axis and Motor connections are 5V. I have not checked all of the M1 outputs but those are the ones that are 3.3V, those signals are just on off types, and like CarbideBob said you only need them above 2.5 volts to change the state from off to on. Now if you want to opto isolate everything you can, but I do not think you need to convert any of the 3.3V to 5V. Guys with more electronics experience please correct if I’m wrong here. I plan on opto isolating any input to the Brain on the M1 and M2 connectors but not the Axis inputs. I hope this helps.

[Jay C]

Jess:

If you look at the output of the Brain on pin 7 on both of the Motor and Glass Scale it is 5V just what you want, no 3.3V, so all of you Axis and Motor connections are 5V. I have not checked all of the M1 outputs but those are the ones that are 3.3V, those signals are just on off types, and like CarbideBob said you only need them above 2.5 volts to change the state from off to on. Now if you want to opto isolate everything you can, but I do not think you need to convert any of the 3.3V to 5V. Guys with more electronics experience please correct if I’m wrong here. I plan on opto isolating any input to the Brain on the M1 and M2 connectors but not the Axis inputs. I hope this helps.

Again, going strictly by the book ... er manual then it says that all I/O is 3.3V. Can you use 5V ... yes, but only because of the high ohmage (28k) in the pullup resistor and the small differential in voltage. Thus 5V - 3.3/28k = a small current of 61uA When the encoder grounds both micro currents flow through the encoder transistor and Brain sees 0V of logic 0. So long story short, you should be fine.

Jay

[sandefuj]

No, no no....the brain outputs 5V on pin 7 not 3.3V. So not all I/O are using 3.3V.

[Dougal]

No, no no....the brain outputs 5V on pin 7 not 3.3V. So not all I/O are using 3.3V.

Exactly.
I'm currently using this 5v feed for my E-stop. The Brain doesn't seem to mind.

[Jay C]

No, no no....the brain outputs 5V on pin 7 not 3.3V. So not all I/O are using 3.3V.

I think you both misunderstood. The I/O is in fact 3.3V (per the user manual and the link that Dougal provided to the SR forum). Yes, there is a 5V supply on the Brain connector that can be used but it's neither an input nor an output. So you are correct that you can use the 5V supply for your limits and switches. I have not seen where it tells what max amperage you can draw from this 5V source though it may just depend on the wall wart you use.

Jay

[LUCKY13]

How many of you guys have Geckos hooked up to the Brain ?


How do you have the Step/Direction & common wires routed ?


Am I understanding right in thinking that the

Common goes to (X axis for example) pin 7 on XA or XM (all though you could pick this one up from any of the A or M pin 7 connectors)

And the Step/Direction goes to M1 pins 2 & 3 ?


Jess

[LUCKY13]

I talked to Bruce about the Opto's on the encoders. He suggested it was to use them on a servo system only and for the motor encoder and not the scales necessarily.


Basically it was for extra protection on the motor encoder where its possible to get into the high voltage of the motor drive. Steppers don't have encoders on the motor and the scales are just powered from the Brain so no need for them there. I guess if the servo/motor encoders had a isolated power supply instead of feeding off the motor drive then it would not be a problem there either. And its not a problem there of course unless the encoder wiring gets linked up with high voltage from a shorting driver or wiring problem.

So in a nut shell it has no need to protect itself from a 5v circuit ( as many already knew and figured anyway) its just for extra protection on the encoder for the servo. I think with proper wiring and a drive that is designed good there would not be much chance of getting high motor voltage on the encoder lines.

I guess its possible for a drive to short like this, IDK myself, has anyone ever seen the driver short out in a way that the high voltage got to the encoder on a servo system ?

I know wiring getting cross shorted could be a cause but anyone building a CNC rig should be considering any possible problems with wiring and run them in a way to be safe with redundant protection. In other words no motor wires bundled with encoder wires plus protective covering. This I would think is normal practice, but people that are new to such things may not realize. This should help everyone understand what chances there taking if no opto's are used.


EDIT - I dont know how damaging noise coming in on a encoder line could be but I guess if noise coming in is enough voltage/power to cause problems then that might be a nother consideration. As far as I know there is probably not enough power from such happenings to cause any shorting, just signal interuptions. But I may be wrong there, someone might want to chime in on that.


Jess

[Al_The_Man]

I suspect the intention for it was that problems may be anticipated where encoders go to the drives and are possibly supply fed from the drive, often with these types of drives, where an encoder signal also goes back to close the loop in the controller the output from the drive is an intercepted and scaled encoder signal.
I would think for economy purposes, the drive can now be a non-intelligent variety and the encoder would go back directly to the controller and be supply fed from it also, making it a virtually self contained circuit.
Al.

[Kent_Norway]

I have been trying to sort out : "The Brain" or the "Smooth stepper" now for to long. I have also tried to read through all these posts but I simply don't have the time....

Is there any one which can confirm if The Brain has a plug in for MACH3 which is now working ?

Also, I have seen that the breakout boards has been mentioned, is there any special requirements for BB to work with the Brain without any problems?

Has any one done any 3D probing with the Brain ?

Have tried to drop them a email but never heard from them, but now I'm ready to get my machine up and running and need to start ordering some parts.. Had basically decided for the smooth stepper but though I should ask before I eventually made my final decision.

K

[DonnieET]

K

There will not be a plug in for Mach software on a CNC brain.

Donnie

[Kent_Norway]

K

There will not be a plug in for Mach software on a CNC brain.

Donnie

OK. I guess I misunderstood the Firmware release V1.0.5 notes then and the users are I guess relying on the SDK SW...

A smooth stepper would be my solution for now then and I'll guess I continue having a eye on the progress in this thread.

Thanks

K