PDA

View Full Version : IH Optical Limit Switches



SCzEngrgGroup
11-30-2008, 10:23 PM
Does anyone know what optos are used in the IH optical limit switches?

Regards,
Ray L.

Runner4404spd
11-30-2008, 11:58 PM
pretty much an optical switch will work. the IH units actually use a small board inside the housing with what looks like a few resistors and the wire connections.

SCzEngrgGroup
12-01-2008, 12:18 AM
pretty much an optical switch will work. the IH units actually use a small board inside the housing with what looks like a few resistors and the wire connections.

Not if you want high accuracy and good repeatability. And, the IH switches use dual optos, so you can have separate limits and home switches in the same package.

Regards,
Ray L.

LUCKY13
12-01-2008, 03:19 AM
HimyKabibble, do you need to replace some ? Didnt you get a set of switchs from a guy on the forum ?


I have a large quantity of switchs but they are a little diff than what I have seen in the IH switch boxs (you can look on BoB Warfields Blog and see the IH switchs) well infact he has pics in one of his threads.


I dont think the ones I have are duel though, but they are so small two could be run side by side. It would take a slightly different trigger arm than IH uses though. The slot on there switch is horizontal to the case to where mine will be vertical.. Really now that I think about it mine could be mounted under the rod/shaft with a simple trigger hanging down off the shaft so it would move through there gap.


The ones I have are 5v with TTL output and the half way test I have done show they would be very repeatable & accurate (although my test where not that good). I could send you a couple of them to test and try if you thought they would do you any good.

Here is a link to a PDF if it helps
http://media.digikey.com/pdf/Data%20Sheets/Honeywell%20Sensing%20&%20Control%20PDFs/HOA7720-HOA7730.pdf (http://media.digikey.com/pdf/Data%20Sheets/Honeywell%20Sensing%20&%20Control%20PDFs/HOA7720-HOA7730.pdf)


These the circuit boards for them is inside of the housing


Jess

SCzEngrgGroup
12-01-2008, 10:38 AM
Jess,

Yes, I bought two of the IH switches from someone here, but I need several more. I want open-collector outputs so I can run higher voltage to improve noise immunity, and I like the idea of a dual opto, so both home and limits can be implemented in a single switch per axis. Also, the IH ones are known to have excellent resolution and repeatability, which is largely a function of the performance of the opto itself, so rather than "reinvent the wheel", I'd like to use the same ones, if I can find out what they are. Unfortunately, they're glued in place, with no visible markings.

Regards,
Ray L.

Cruiser
12-01-2008, 12:46 PM
SO Have you contacted Gene or his boys ? Wouldn't that be where ya need to be inquiring ?

SCzEngrgGroup
12-01-2008, 01:28 PM
SO Have you contacted Gene or his boys ? Wouldn't that be where ya need to be inquiring ?

I have done that. Of course, they may not want to say....

Regards,
Ray L.

LUCKY13
12-01-2008, 03:06 PM
Is the IH a dual switch ?


I really didn't realize just how many different types there are. I will have to go through my collection ( I have a Hugh amount of them but I bet dual channel is going to be thin).


Not sure but is this something like you need ? Although I cant find any that is pre wired, all would have to be put on a board for wiring.


http://www.components.omron.com/components/web/pdflib.nsf/0/C99E746081A1A42985257201007DD5D9/$file/D21EESX10310305.pdf



Jess

SCzEngrgGroup
12-01-2008, 04:31 PM
Is the IH a dual switch ?


I really didn't realize just how many different types there are. I will have to go through my collection ( I have a Hugh amount of them but I bet dual channel is going to be thin).


Not sure but is this something like you need ? Although I cant find any that is pre wired, all would have to be put on a board for wiring.


http://www.components.omron.com/components/web/pdflib.nsf/0/C99E746081A1A42985257201007DD5D9//D21EESX10310305.pdf



Jess

Yup, the IH switches are dualies, which is a nice feature. I looked at that one you pointed to, but my concern is it does not have a terribly sharp cutoff (look at the current vs shutter position graph). Some others have a much sharper cutoff. This can be somewhat mitigated by proper biasing, and feeding the output through a Schmitt gate, but I have no feel for just how well it would work.

Regards,
Ray L.

SCzEngrgGroup
12-01-2008, 10:17 PM
Oops! I took another look, and the optos are not dualies after all. It is two singles right next to each other. I've decided to go with the Omrom OPB941. It's got a small aperture, Schmitt trigger, open collector output. $7 each from DigiKey.

Regards,
Ray L.

LUCKY13
12-02-2008, 10:04 AM
Shew, LOL, you went straight over my head. OK so your saying there is a big difference in how fast the sensor can react, and I guess how sharp of a switching action it can have ( thin line that defines on/off).



I guess I kinda knew this but thought that for the most part they would be repeatable even if such things are true.



I would be very interested in seeing how you put two sensors together and how these are hooked up and used with a controller. I am afraid when it comes to electronics I am starting from zero.


I remember Aaron have a test rig setup to where he had the switches setup with a micrometer mounted on a piece of aluminum. His setup worked so good that when he got on the edge of triggering he could lay the assy down and it would un trigger just from the temp change from where he was handling it. Then without moving the micrometer he could touch the assy and the thermal change would trigger the switch. I guess he knew what he was doing & how to chose the right switch (sounds like you do also).

I found a link to his article of testing the limit switches.

http://web.archive.org/web/20050211085539/http://www.industrialhobbies.com/


Just go to the left side of the page, near the bottom, and the link to the article is there.


Jess

SCzEngrgGroup
12-02-2008, 10:42 AM
Jess,

Optos are not like mechanical switches - always either on or off. They have a range in-between where they are neither fully off, nor fully on. A run of the mill opto may allow you to move the flag as much as 2-3 mm before it transitions from fully off to fully on. During that distance, the PC may see it switch on and off several times, due to noise pushing the level just over the threshold. If you test it 10 times, you'll find the PC sees it switch at slightly different positions each time - not good. Getting the kind of repeatability IH does depends very much on picking the *right* opto (one with a very small "aperture", and a sharp cutoff characteristic, preferably one with a Schmitt gate in it to add hysteresis), and properly biasing it (setting the optimum LED current, picking the correct phototransistor load resistance, etc.) , so that range of uncertainty is minimized. The mechanical design also has to be very robust.

On mine, one switch will serve as the home switch, and be wired directly into Mach. The other will be the limit switch, which will be wired directly to the servo controller, and cut AC power to the servo power supply if the limit is hit. This will, of course, necessitate a manual over-ride to allow the system to be powered to back it off the limits. But, in actual use, these limits should *never* be hit.

Regards,
Ray L.

LUCKY13
12-02-2008, 03:26 PM
Good deal, I can see how the higher voltage will help over come noise also.


I guess learning these little things can really effect the outcome of your system. I will start out with the best of the switches I have, and learn from them, then upgrade from there..




BTW, I wanted to mention about your motor mounts. You stated you might be interested in making a few for other people(to sale). I am not sure if they will work with my mill, but I might be interested in some here in a few weeks.


I am doing some things a little different on my motor mounts though but I am guessing your mounts could be setup to work with what I am doing.


Mainly on the Y axis, I am going to bring the motor down and then mount the motor from the back of the mount. This would have the motor hanging down, and going back under the base of the mill ( its a bench top much like the IH).


The bench I am building will be built so that under the mill will be open right between the mounting bolts (at least about a 1ft tall hollow). This will give a place for the coolant to settle and drain, plus area for the motor. I am disabled and I need the mill as low as I can, plus up to the front edge of the bench as far as it will go so that I can reach it from my chair. I will build a door/panel so it can be removed to get to this area when I need to service the motor or the drain. Then build seals into this door/panel so it holds the coolant but it should not backup no more than a inch deep or so ( if so I build bigger drain).


Anyway I think you have the mounts setup for a longer belt than I may be needing ( not that far along yet to know) . I expect they may be able to have changes done to them. I am in no hurry ( that is if your even interested) and I do still have specs to be figured before I know what I need. If you are interested we can work on it through the next few weeks.

I may be going in for a operation before the end of the year,
so I'm not 100% sure when I will get figured out what I do need. I go to the Doc tomorrow and its hard to tell what they will decide to do ( if I even let them do anything). But this could go from as much as a full string of operations, down to maybe just a out patent type deal, or even a combonation of both. I hessitate to let them do much from past expeirances with other ops, at least till the need gets to a point of no choice. For now I will just see what they come up with and go from there.


Jess

SCzEngrgGroup
12-02-2008, 06:49 PM
Good deal, I can see how the higher voltage will help over come noise also.


I guess learning these little things can really effect the outcome of your system. I will start out with the best of the switches I have, and learn from them, then upgrade from there..




BTW, I wanted to mention about your motor mounts. You stated you might be interested in making a few for other people(to sale). I am not sure if they will work with my mill, but I might be interested in some here in a few weeks.


I am doing some things a little different on my motor mounts though but I am guessing your mounts could be setup to work with what I am doing.


Mainly on the Y axis, I am going to bring the motor down and then mount the motor from the back of the mount. This would have the motor hanging down, and going back under the base of the mill ( its a bench top much like the IH).


The bench I am building will be built so that under the mill will be open right between the mounting bolts (at least about a 1ft tall hollow). This will give a place for the coolant to settle and drain, plus area for the motor. I am disabled and I need the mill as low as I can, plus up to the front edge of the bench as far as it will go so that I can reach it from my chair. I will build a door/panel so it can be removed to get to this area when I need to service the motor or the drain. Then build seals into this door/panel so it holds the coolant but it should not backup no more than a inch deep or so ( if so I build bigger drain).


Anyway I think you have the mounts setup for a longer belt than I may be needing ( not that far along yet to know) . I expect they may be able to have changes done to them. I am in no hurry ( that is if your even interested) and I do still have specs to be figured before I know what I need. If you are interested we can work on it through the next few weeks.

I may be going in for a operation before the end of the year,
so I'm not 100% sure when I will get figured out what I do need. I go to the Doc tomorrow and its hard to tell what they will decide to do ( if I even let them do anything). But this could go from as much as a full string of operations, down to maybe just a out patent type deal, or even a combonation of both. I hessitate to let them do much from past expeirances with other ops, at least till the need gets to a point of no choice. For now I will just see what they come up with and go from there.


Jess

Jess,

You need to be careful making the belt too short. The limiting factor is the number of teeth engaged on the small pulley. You must always have at least about 6 teeth engaged, depending on the type of belt you use. With a reasonable reduction ratio (mine are 2.5:1 for X/Y, 4.8:1 for Z), you can't make it very much shorter than I did. You'll notice most commercial drives use even longer belts than I did. If you go to www.sdp-si.com you can download PDFs of their belt drive design guides, or sign up to have them mail you the hardcopy for free. All the information you need for properly choosing the right belt, pulleys, etc. is in there. There is also a nice web-based tool on their website for designing belt drives. It lets you enter the pulleys you want to use, and the number of belt teeth, and it will tell you the center-to-center distance for the pulleys, the number of teeth engaged on each pulley, drive ratio, etc.

Regards,
Ray L.

LUCKY13
12-03-2008, 02:56 PM
Understand.


What is the center to center of your screw/shaft combo ?


Looking at the pics it appears to be very wide. I am guessing the pics are effecting the way it looks, but judging by the pics it appears to be somewhere around 12" center to center. This must not be the case and the pics are distorting the appearance.


If shorter spacing is needed than the belt/gear designs call for then could using such methods as adding a idler make it work. Backlash and response to me would be the worry in such a setup, yet idlers are used in many belt/gear setups so there may not be any bad issues.


BTW I would like to thank you for your time & info, most seem to be a little to busy to discuss design and help figure the need.

Like your limit switch problem. My own limit switch setup will end up being much better now after going over the info. It may take me some time before I can improve on the ones I will start out with, but I know what to look for when buying a switch now to replace them. Or I may find I have better switch's in my collection after I go though them. I find this CNC stuff forces me to learn through things I have no background with, not a bad thing, but one that takes me some time to take in.



On the gearing for my mill, I am thinking of ordering a 4 to 1 setup for testing. If I can achive anywhere from 175IMP upto 250 I should be good. Infact I don tthink my little benchtop mill will handle any more than this. If it doesnt work out I will use it for the Z and order what ever is needed for the X & Y.

With my controler I will have 500Khz's available instead of the normal 45,000hz that Mach3 offers. I dont know what SmoothStepper is able to produce but I feel like its possible for me to drive my motors at max RPM with the controler I will be running ( CNC Brain ). Its a USB based system much like the SmoothStepper, but I think it will flow alot more info. Plus its not dependent on the computors CPU to do the work or calulations, so I am kinda expecting very good performance. We will see.


Jess

SCzEngrgGroup
12-03-2008, 03:55 PM
6.9", center to center. The large pulley is 70 teeth, the small one 28. These are GT2 belts, not XL. My Z axis uses XL belts, same center spacing, but 72 teeth on the large pulley, 15 on the small one.

I think aiming for anything over about 100-150IPM on a small machine is rather pointless. It's good for bragging rights, but has little practical value, unless perhaps you're doing a lot of 3D profiling, and creates some real risks. On a small machine, you can run end-to-end very quickly, which means if anything does go wrong, the chance of doing some serious damage when it runs out of travel is very high. The faster you go, the more room you have to provide for it to "coast down" before hitting a hard stop, and that distance comes right out of your work envelope. My knee mill has a range of 34" in X, 14 in Y, 16 in Z, and X and Y are easily capable of 400 IPM, but I run at 150, because things happen much too quickly at 400. At 400 IPM, I can run end-to-end on the X axis in barely 5 seconds - not a lot of time to react. A moments inattention while entering MDI commands can do a lot of damage. In terms of program run-time saved, the higher speeds *might* buy be a minute or two per hour - hardly worth the risk.

SmoothStepper will run up to 4Mhz step rates, which is WAY beyond what there's any practical use for. I limit mine to 128kHz. The load on the PC is minimal, as the SS does all the "heavy lifting".

Are you using steppers or servos? I assume you must be planning on running double-nut ballscrews?

Regards,
Ray L.

LUCKY13
12-04-2008, 03:54 PM
Yes double nut ballscrews all around. I am even going for the bigger screws (.750 x .200). Many people will run the 5/8th screw instead of the 3/4, heck if I think I can fit a 1" it will get them ( will size them up when I tear the mill back down).


The motors will be the same exact motors your running (850 servos @ 70v 4200 rpm) The power supply is 72v @ 20 amps (1500watts). Thats why I was so surprised when I seen what they where performing like on your mill.
Then to think the IH Industrial version runs a 1124oz-in servos.

I do need to push this mill because of the smaller high speed stuff I will be doing. But it will also be milling steal. The base & column will have epoxy work to add strength. It will gain about 150 to 200lbs of weight,plus the added stiffness. The bench I am building for it will be way over built so that when I bolt this mill down, it will be solid. I am trying to over build the whole setup, that is one reason I ordered the CNC Brain controller. It takes the PC out of the calculations and will feed the drivers much better/faster. Plus it can have closed loop control which should help it run past the mechanical ability of the machine and still stay on track with no lost steps. How well it does this I cant say for sure myself, when I get it in I will push this stuff to limits even if I don't intend to run there all the time. I like to know the limits of the complete setup so I know how much cushion to allow for safe milling without loosing every other part because of lost steps. If I can get this thing right, that may be more of a self impossed safty limit I deside on instead of what limit the machine can really mill at. Lots to find out and learn before I even get that far though.



Jess

SCzEngrgGroup
12-04-2008, 06:05 PM
I'd suggest you take a good look at the SmoothStepper as well. It also off-loads all the heavy computation from the PC, and it's much cheaper than the CNC brain, and very well supported. Mine has worked flawlessly from day one.

"which should help it run past the mechanical ability of the machine" - Sorry, but that's just not possible, and it would be poor practice in any case. And if you did, you'd be eating tools all day long. You will find yourself limited by spindle power, and machine rigidity, unless you either do a poor job designing the conversion, or push it beyond what is prudent. My drives are *never* the limiting factor when I push my machine. First is spindle power (and my spindle is 3HP), second is tool breakage (I can easily snap 1/2" tools without losing position), third is machine stiffness (which limits accuracy, finish quality and tool life). And if you're going for a really high-performance machine, 3/4" ballscrews won't get you there. I have 1" Nook XPR screws on mine, and they are surprisingly "springy" when pushed. Big machines use BIG screws - like 1.5-2" or more. You will be amazed at how much flex you'll find in every corner of your machine once you start pushing it. You'll need to completely re-design *all* of the bearing mounts, the nut mounts, and on and on.... Getting a machine to 0.001" accuracy is not too difficult. Getting it beyond that gets exponentially more difficult and expensive.

Regards,
Ray L.

LUCKY13
12-07-2008, 12:09 AM
I'd suggest you take a good look at the SmoothStepper as well. It also off-loads all the heavy computation from the PC, and it's much cheaper than the CNC brain, and very well supported. Mine has worked flawlessly from day one.

"which should help it run past the mechanical ability of the machine" - Sorry, but that's just not possible, and it would be poor practice in any case. And if you did, you'd be eating tools all day long. You will find yourself limited by spindle power, and machine rigidity, unless you either do a poor job designing the conversion, or push it beyond what is prudent. My drives are *never* the limiting factor when I push my machine. First is spindle power (and my spindle is 3HP), second is tool breakage (I can easily snap 1/2" tools without losing position), third is machine stiffness (which limits accuracy, finish quality and tool life). And if you're going for a really high-performance machine, 3/4" ballscrews won't get you there. I have 1" Nook XPR screws on mine, and they are surprisingly "springy" when pushed. Big machines use BIG screws - like 1.5-2" or more. You will be amazed at how much flex you'll find in every corner of your machine once you start pushing it. You'll need to completely re-design *all* of the bearing mounts, the nut mounts, and on and on.... Getting a machine to 0.001" accuracy is not too difficult. Getting it beyond that gets exponentially more difficult and expensive.

Regards,
Ray L.



HimyKabibble, I looked unto the SS & Mach3 very hard before I bought the CNC Brain. Although the SS does do somethings that are related to how the CNC Brain operates, there is much more to the Brain than this.

Mach3 & SS are very well supported and this may be something I do regret in the beginning. Lets face it, Mach3 is a very well matured system that has many support avenues with lots of well documented features. It was hard for me as a beginner to walk away from that, and I thought long and hard before I did. But I tried to look at this in a
" where will I be a year from now kinda way". In that respect I believe I will be better off even though I will be behind for a while until the Brain starts maturing. All of the Plug-ins and scripts many have worked on with Mach I will be without for some time.

The CNC Brain is maturing at a much faster rate than other systems in the past. The guys behind this system really have put it together in a way that will make growing/expanding much easier than other systems. Scripts that have been written for Mach, could very well just transfer over without much rework. Being a beginner at this CNC stuff I may very well find out deferent later when I put this system to use. But IMHO I think many will be looking into this system as time goes by and it starts maturing.

One thing I seen very quick about the pricing, by the time I purchased Mach3, SS and few Plug-ins there was not much difference between the two in pricing. Now when you add closed loop to the CNC Brain the expense of the glass scales will add to the cost, but I feel like it will be worth this extra expense for the ability to have a true closed loop system.


As far as Machining past mechanical abilities, that was probably not the best way to describe my meaning. A bench top mill will have many limits with out a doute. I hope to address each one and improve each the best I can to increase the ability of the mill. The Brain want make any mill a super mill, but it can help over come many problems. Like a spindle bogging down for instance, it can slow feeds down so that they match the lower RPM the spindle has fallen to. It can keep the machine from loosing steps when it encounters vibration/resonance that would cause lost steps. A Bench Top mill would probably have more trouble with such vibrations, as say compared to a Bridgeport which is much more ridged. IN this respect, yes it could help go beyond the machines normal abilities if it takes you past a point that your machine would normally loose steps, and corrects these mistakes to keep the machine on track. Of course this is a moot point if this level of losing steps is beyond what the spindle can keep up with, unless it is during a simple rapid move,tool change or probing operation where speed could really help.


In the end if I can get this machine to just not be a dog on its speed, and not be throwing away parts every other try because of control problems, I will feel like I succeeded. Speed is something I am very familiar with and I always go 150,155,160,165mph,ect,ect,ect.... before I attempt to go 200mph. :) Safety and accuracy will come first. Problems that show up, I will have to deal with them as I learn about them. Stronger mounts, biggest screws, upgrades to the spindle, adding ridgitity and many other things I am trying to consider as much as possible and do what I can to help there ability. At what level these upgrades will bring the machine to will have to be seen with the machine as a whole. As aleady meantioned with all machines, the weakest link will be the limiting factor.



Thanks Jess