What is the general concensus on using a breakout board with an IH CNC? Is there any reason (other than saving some cash) not too? Seems like there are many good reason to do so.

Thanks

Anyone here using this or any breakout board? I am leaning towards one with my kit arrive tomorrow but would like to know if there are any gotchas to be aware of when using the board from Bob.

Thanks

I used the CNC4PC board with mixed results. The first 1 had to go back because it didn't provide enough current to drive the Geckos all the time. The second one would only work on certain outputs, but it's working fine now. Don't know anything about the Bob Campbell board though, sorry. You definitely need a breakout board though, 1 to make the connections easier and 2 to protect your computer.

Thanks for pointing out the CNC4PC board - I had not seen that one before. Quite a price difference between it and the Campbell board.

i use two campbell boards and one of his relay boards with my ih cnc. can't recommend him enough. excellent product and... you do get what you pay for!

That's what I needed to hear - got one ordered up.

Thanks

I have two PMDX-122 boards that I've been using for about a year without problems of any kind. Another user mentioned that his VFD created enough electrical noise when his PC was either off or not running Mach 3 to cause the outputs on his PMDX-122 to change state randomly. A suggested work-around is to add pull-up resistors to the input lines on the board (which would cost about $1.00 and take about five minutes if one end of each resistors were simply inserted into the wire terminals of a signal line and the other ends of all the resistors tied together and connected to a 5VDC line from the PMDX board).

Richards brings up an interesting thought. Be really careful around a CNC machine. Don't have the drivers powered up if there is no PC in charge. A lot of these breakout boards have a safety charge pump circuit for that purpose.

Alternatively, it isn't that hard to wire a circuit that will not apply power to your motors until your PC or other front end controller (mine is a GRex) has had time to boot up. Shut down said PC or front end, and the power goes off for the motors.

I used a delay-on-make relay for my circuit. This is a handy little relay that waits a specified amount of time after it "sees" lline voltage before letting it through. You can set the interval from a few seconds up to minutes. That relay gets the line voltage going into my GRex (or your PC), waits long enough for it to boot, and then sends it on to power up the motor DC supply.

Be sure if you are going to cut power you do it in the right way so as not to blow your Geckos--you want to just cut the power to the DC supply. The filter capacitor will let it down gently. Don't just stick a relay (or a fuse for that matter) between your Geckos and the motors.

I've also wired a hard E-Stop (actually a couple) between this delay on make relay and the line voltage. If I hit one, it also pulls the power to the motors. Lastly, my spindle gets its juice here too. I'm gonna be hitting those E-Stops any time I have to reach my hands into that machine to change cutters or whatever--I want to know without any doubts there is no juice to those motors.

Just a thought!

Best,

BW

I started wiring an IH CNC kit to a Campbell board and wondered if someone would explain how to power the limit switches. Aaron's wiring diagram calls for 5V but I don't see a dedicated 5V supply on the board for this. Perhaps I can use the 5V that feeds the geckos? Alternatively perhaps I could put a resister to cut the voltage down... What do you think?

Thanks

I'm extremely conservative when it comes to hooking up devices. In my prior life, I designed, programmed and installed process control computers in the photo industry to run the Kodak-S series printers found in most professional photo labs (before the digital photo revolution made my services as valuable as a buggy whip manufacturer). I always use opto-isolators whenever interfacing anything into a control board. That means that the limit switches would have their own power supply and that the only connection between them and the controller board (breakout board) would be light pulses. The inexpensive Texas instruments TIL-111 series of opto-isolators can usually be purchased for a few cents each. My favorite is the Opto-22 G4 modules that run about $15 each for input modules, although they require a 'rack' for another $70 or so.

Connecting signal lines on a machine that has stepper/servo motors, VFDs and assorted other mechanical/electrical devices directly to a controller board is just asking for problems, either now, or sometime down the road. An extra $10 to $250 dollars up front to do things right will keep you safe and sane.

Some breakout boards are already opto-isolated. (I'm not familiar with the Campbell board - which might work perfectly right out of the box.) If that's the case, then you would only have to add a power supply that is independant of the controller's power supply to keep things isolated.

Anyone know how to interface the Geckos so that a fault causes Mach3 to stop? The behavior currently is that the Gecko faults but Mach thinks the moter is still moving.

I'm assuming by "fault" you mean a servo fault and you are using a Gecko 320 to drive it. The doc for that drive (http://www.geckodrive.com/photos/G320_REV-4_MANUAL.pdf) discusses how to access its ERROR output and still allow the drive to be reset via a switch. If you've got an output, and enough inputs on your breakout card, you can presumably send the result back to Mach and set up Mach to shut things down on a fault, much as it would if you hit a limit switch. This will require a bit of macro work as well as a bit of electronics, so you'd best get some help on the Mach and Gecko support boards if you really want to pursue the option.

Best,

BW

http://www.aree.com/gecko/err.html

i've got this one and i'm running it with the cnc4pc board.

Runner - That board looks like just the thing! I am curious about the recommended use of relays that the board's documentation suggests. I thought I had read that abruptly killing the voltage to the geckos can be damaging and hence the power supplies have a cap/bleed off resistor for cushion when power is lost.

I started hooking the IH limit switches up tonight to a Sound Logic breakout board and can only manage to get either the home or the limit on each axis to trigger. There are four wires coming from each limit switch: common, power, sensor 1, ans sensor 2. Inside the limit switch two Fairchild H22LTB optical modules are used. I first attached the power, common, and sensor 1 to the home pins. After configuring Mach3 a little I was able to trigger home. I then attached sensor 2 lead to a set of limit signal pin leaving the power and common connections free on the limit pins. After doing that neither limit could be triggered in Mach3. Detaching the home sensor lead now allowed the limit to work. I also tried attaching both sensors to the home through diodes without success. In short I only seem to be able to attach only a single limit switch on each axis. Anyone have an idea of what might be happening here?

Thanks

Well, after more tinkering I managed to get it to work and have sent what I did to Cullins for him to verify. Basically I added a diode to each side of the limit switch, tied the other end of the diodes together to the Sig line on the limit input to the board, and added a resister between Sig and Gnd to force a natively low level. It seems the Sound Logic board expects the limit switches to sink the Sig current to low when not triggered and source it high when triggered. When both sides of the limit switch are connected to the single input on the board they sink each other low and a high value is never seen. By adding the diodes they don't sink each other but the normal (unconnected) level and the level when the limit switch is tripped are not very different (but each side of the switch is distinguishable). Adding the resister sets the default level low and each side of the limit switch is easily distinguished. It now works and so far no smoke has been seen :)

Wildcat, This is some very good info that i will appreciate quite soon, It sounds exactly like what Aaron had me do with them anyway, but in tieing it in to "P" port signal. Allthough the opt limits don't need the breakout and neither do the Gecko's, I am opting for the cambell combo board as well just to clean things up a bit and link with my L200 vfd and coolant pump. I tried to do it with the cnc4pc spindle and relay, but have nothing good to say (period). My thought here is I'd like to see a schematic of this hookup of yours because i know for a fact i'll need the same for my machine ! I think I can see what your saying but when it comes to electronics, A picture is worth a lot ! and Information is Gold ! I may be great at mechanical and alright at machining (my vocation) But not very advanced (elementry) with electronics. I just want to get this machine solid so i can sit down and spend some time with turbocad/cam 12 and learn how to get along with it !

Hey Guys;
Just a quick peek at the above for this fast moving friday, but did I miss something?? What happened to the old fashion charge pump as a system monitor? They are inexpensive and will not allow the system to fire (relay lockout) unless mach is outputing a solid freq on pin 17. I use them on all my systems and when I have a computer problem, the system HV drops out a as a safety and the motors stop. If this has been addressed, just disregard.
Bill
:)

Bill, charge pump is part of the board but not the discussion !

Cruiser - I never made a full schematic. Actually I am still modifying my electronics but below is how I interfaced the IH limit switches to the Campbell board. This "circuit" must be replicated for each limit so you will need six diodes and 3 resisters. You can interface this to either the home or the limit pins on the breakout board - they are tied together on the board. The exact diode is not too important - I used 1N4006 since I had six of those. Aaron sends three 1n4004 with the kit so if you can find three more... I chose a 270 ohm resister as it provided a nice differential between low and high and kept low to less than a volt. Also, I tied all the shields together and grounded them. Also, for some reason, if you wire the Geckos as Aaron has in his schematic and then tie the Geckos to the breakout board the motors turn the "wrong direction." You may need to reverse the leds and encoder phase connections. Please clarify if you need a full schematic and I will do a better job drawing one up.

Wildcat, This is exactly what i have been trying to get a hold of. I kinda sorta know what you are saying but i'd still really like some sorta sketch to clarify, I would only need a sketch on this portion of the hookup, the rest of it is covered pretty good. I suppose i could study this and figure it out, but i'm not electronic to any comfortable degree. I just received my combo board yesterday and printed out the info, stored the etc...........can you please do a quick sketch for me ? I think culluns forgot about my request for info, and you are the only source for now. drop me a pm and i'll send my addy ! ok ? And Thanks

Cruiser - So sorry - I posted the diagram and forgot to include the link. Cullins is apparently out this week visiting a customer (have been trying to get information on the spindle control board). The breakout board is a great product but the customer support is a little lacking.

http://www.cnczone.com/gallery/data/500/medium/IH_Lim_SLB.jpg

Wildcat, Thank you, as they say a pic is worth a thousand words, or so. one look at your sketch and I see it all, quite simple really Again thankyou, i'll probably start on it friday morning, hows yours coming along so far ?

Crusier - I placed a more complete diagram in my photos. I am not really sure how to add a link to the fullsize version but if you check the photos uploaded today you will see it. Mine is going but I want to add a Err/Res controller and big red stop buttons yet. Ordered a VFD and motor today but sounds like the spindle speed control board that interfaces with the breakout board is a ways out yet.

http://www.cnczone.com/gallery/data/500/medium/Layout2.jpg

Wildcat, yeppers, that is what i saw too and that is why i ordered the combo board to replace the poco cnc4pc boards i bought to get the vfd on line. I may make someone a great deal on the cnc4pc stuff as this combo board looks sweet and has it all. actually the cnc4pc probably would have worked for me if i could have received some info on how to hook up the drop res to the sig's, as it was it only partially worked and the vfd knocked out the limits somehow, not so totally isolated i thinks. have a good one !

My old eyes can't see well enough to be sure what you've posted with the scematics; but, if what I think I saw is correct, I would like to make a few suggestions:

1. Put an opto-coupler between each 12V sensor and each 5V input. My old rule of thumb is that you don't mix voltages. Why take a chance on the diode shorting out and frying the rest of the circuitry?

2. One device per input. One of the first things that I learned when I started designing was that it is never a good idea to have more than one signal per input. That's one of the reasons that they make multi-gate TTL.

I've been testing some proximity sensors that work off of a 12V power supply. I let the proximity sensors drive the photo-diode on a TIL-111 type opto-coupler through a 1K resistor, which gives me about 10ma through the photo-diode - a current that is about what I want. The other side of the opto-coupler is 5V. The collector on the photo-detector is connected to a TTL input line. It's very simple. It doesn't cost more than a few cents per input. It is robust. It doesn't break the rules.

Richards - These are all great ideas but I am not sure why this would be needed in this particular case... there is an opto isolator in the Sound Logic board between the limit switches (which themselves are opto isolators of sort) and the computer. The IH limit switches can take a range of voltages and 12v is fine. Also a similar signal tying is performed in Sound Logic board between the Home and Limit switches and is used by the IH wiring diagram (in particular the "Alterantive" wiring diagram). Did I misunderstand something?

There is no 'absolute' right or wrong way to build an electronic circuit. Some would say that if it works, then it is right. I prefer to follow other standards. As I posted above, my inputs only allow one signal line. (However, one output line can drive multiple inputs - the old Fan-In/Fan-Out principle.) Steering diodes usually work very well, but I don't design that way. In my world, each voltage level lives in its own space. A 12V circuit only lives in a 12V space. When the signal needs to go to a 5V space, it has to pass through an opto-coupler. As far as I'm concerned, multiple opto-couplers are acceptable as long as each voltage space has its own power supply. Overly cautious? Perhaps. But it works.

Richards - The 12V from the limit switches pass through a opto isolator/coupler on the Sound Logic board before enter the 5V space of the computer. And yes, there is isolation of the 5V and 12V supplies on the board. I believe this is mentioned in the Sound Logic documentation and it is clearly shown in the schematic of the Home/Limit portion of the board. Am I missing your point?

I've never uploaded a schematic to this forum, so I've got my fingers crossed that this will work. In any case, the partial schematic shows a $10 circuit using a microcontroller that I'm using as my 'glue' chip to interface I/O to a breakout board. Since it is a microcontroller, it can flash the LEDs, ring bells, sound buzzers, send messages via a modem - well, maybe that's a bit much. My point is that a very simple circuit can make life easier without violating standard practices.

Edit: It looks like I forgot to label the lines going to the breakout board. P3.5 is the signal line going to the breakout board. The grounds between the breakout board and the microcontroller are connected. NOTE: The 12V supply and 12 ground are NOT connected to the breakout board.

FWIW: Alan has a new Gecko Err/Res controller board (was mentioned earlier in this thread) that includes the relays and power supply. In case there is any interest the URL is

http://www.aree.com/gecko/new/

Wildcat, How is your system limits working ? I wired mine up and its not quite there with 270 ohm, i'm wondering what's differant. It appears close but only pulled down to the 3v range !

Cruiser - They are working quite well. I believe the voltage range that I see is just under 1v when low (limit not tripped) and nearly 10v when high (limit is tripped). The limits are being powered with 12v. Which board are you using? Is it possible that you have the diodes backwards? Are you powering the limits with 5v or 12v? You want current to flow across the diode from the limit side to the breakout board side. Have you measured the low and high values for a limit before the diode?

Well, I have figured out what was wrong with my system ! apparently with my system the 270ohm res was holding the voltage too low to trigger the optoisolator. So, I got a 330ohm res, UNT VIOLA ! IT WORKS FLAWLESSLY ! SO ..... add the 330 resistor into the works and a note as i will forget it all as soon as the residual heady ach goes away. Geez, it took me too long to realize this dilama maybe that is why i'm a mechanic and a machinist and everything else butt not an electocutionist ! Now i must wire up the combo board to my hitachi L200 & I am feeling fear of another heady ach ! has any one got me some nerve relieving guidelines to follow here ? PLEASE ?

Cruiser - Good deal on getting the limits working. I'll remember if I see any problems on the limits that the resisters might need to be changed out. Regarding the VFD, I am just starting down that path. Received a prototype board from Cullins recently and just got the VFD and motor working (very cool) now to get them all working together...

Been thinking about an enclosure... on the cheap. Have you given an enclosure any thought? I have a working bench and planned to slip some sort of tray, like on Aaron's site, made of wood and plastic lining. With the new motor chips are being flung a long ways!

My enclosure has been multi posted in several places i think, starting in the old thread bout the tormac enclosure wanna build one, mine has progressed some, got some lexan in but still some more to get done and cover the lectronics which will be all visable from front of machine, i'll post a new pic soon when it is considered done and on line maybe after this weekend i hope. then it be time to sitts down and learn turbocad/cam to some degree. i have several projects that will need extensive cad design work.

Cruiser - I got the VFD working today. I have an SJ200 which is really similar to the one that you have. I switched the VFD to sink mode, used pins 1 and 2 for forward and reverse, L for direction GND, and pins O and L for speed and GND. I am using the prototype daughter board for the Sound Logic Breakout Plus. Oh, and I had to set A001 and A002 both to 1. Be sure to put your VFD on frequency display (D001 I think) and try decreasing the speed from 3600RPM and make certain that the Hz are dropping. I had to do a little adjustment to get speed adjustment over the entire range.

it sounds just like mine for connections, i've run mine before on the cnc4pc but was inundated with noise so i unhooked it and run manual till i got the combo board, i like the idea of the optoisolators, i'll wire it up on friday and see if wee ! or pee !

I just thought i'd throw this in as i forgot to mention it in the post where i corrected problem, What i found was that i had two optical limits switchs on one circuit and only one was tripping mach but both were tripping the led's! I'll admit it took me too darned long to logic this out but here is the deal ! the voltage differential between the working and faulting limits was .024 volts ! I stalled on this for a while till i rememberd that the optoisolator is a switch and requires the proper imputous to trigger ! that is when the bells and whistles went off loud and clear so i ran to the Shack and got the 330 ohm resistors. Why the experts that sell this product didn't offer this as something to check is beyond me. I felt as tho i was dealing with salesmen that didn't really know there product ! At any rate this is a done deal and this morning i'm going to go out and wire up the VFD and if that works then my conversion is a done deal and "ON LINE!"

Well, I wired up the VFD to the Combo board friday. It didn't work for a long while, then i found that the jpi jumper was not ! shipped on the board. I scrounged one up and got some function. But i don't have reverse spindle. the rest of it is close and will need a little fine tuning, but i want my reverse and i can't wrap my old mind around this one yet. I wait to see what Robert has to say, or anyone else for that matter.

Cruiser (and all that follow),

For reference here are the resister values that I tried and the corresponding differentials. However, I am using the Sound Logic Breakout Board Plus so perhaps that is where our differences are coming from? I am switching it out for the Sound Logic Combo Board though to get better VFD and relay control.

Ohm V Low V High
2.2k 4.55 9.58
470 1.6 9.1
270 .988 8.28
220 .826 7.00
120 .465 3.77

So is the Cambell Combo the best/ most reliable choice for a conversion (for Gecko 203v's)?

"BEST" ? This is a reletive term & one must concider the system one is assembling & the componants used. The one thing i can say is that in my instance it was a waste of money except that it does drive my vfd quite well. The short fall with my system is that I have been inundated with noise (read stray voltage spikes) that would bring me to tears. I have reverted my system back to basic or direct port wired. and am having no more noise complications. So many use it without any trouble that i guess you'd have to check with them and see what they are using in their system to judge if it will work with what you intend to do !

I use the Campbell breakout board plus with no problems but I don't have a vfd yet. I have a friend who is running 3 machines with the cnc4pc board + vfd in the cabinet with the board and has no problems.

Cruiser,

Is the "noise" coming from your vfd? I seen to recall reading in a previous post vfd's were causing some problems. Are/ were you using shielded cabling?

Shepard,

Thanks for the reply. Doesn't your board have more options than your friend's? I'm kinda leaning toward the Cambell unless someone (who has used it) steers me in another direction.

Cruiser's "noise" problems are a concern but if they are from the vfd that won't be an issue for me since I'm not using one- at least for now.

Thanks

I'm using the Campbell breakout board plus not the combo board and there is no onboard spindle control. I know the cnc4pc board has spindle control and if memory serves it has a lot of features for half the price of the combo board. I don't know about the combo board but home/limit is a pain with the breakout board plus. I originally went with the Campbell board because it supplies the 5volts for the geckos.
I'm sure everyone's experience is different but I was never able to get ANY customer service after numerous emails to Campbell and they have no phone number. Anyway, I'm planning the cnc conversion of my lathe and will be using the cnc4pc board. They have answered my email questions within 24 hours and they run at my friend's house 8-10 hours a day.

Like Cruiser said, the "best" board depends on your application but if you're not running a vfd or need alot of I/Os there are plenty of less expensive boards that will work.

Thanks Shepard,

The lack of support from Cambell and responsiveness of CNC4PC cinches it for me. I ordered the C11G (G is for Gecko) last night. It is less than half the price of the Cambell board and looks like it will do everything I need.


Thanks again,

HB

Did you order a power supply for the card? It does not have one on-board. You need two separate supplies for isolation.

If your VFD has a RUN logic signal (low current DC) you will have to use the Solid state relays to turn on a conventional dry contacts relay. Solid state relays won't work with logic level DC.

I have a 5/ 12 volt supply aleady in the machine and I have a computer 5/ 12 power supply I can use as the second supply. Thanks for the headsup Torchead, I had not thought about needing a second power supply for isolation.

I have a 5/ 12 volt supply aleady in the machine and I have a computer 5/ 12 power supply I can use as the second supply. Thanks for the headsup Torchead, I had not thought about needing a second power supply for isolation.


Computer power supplies typically have a common ground (chassis) and the safety ground of the AC.....That will tie everything together and defeat any optoisolation advantages for ground loop prevention or ground conducted noise issues.

The whole grounding thing gets confusing because no two grounds more than a few inches apart are at the same potential. (ground is not Ground!) The more current there is between the systems and the higher the frequencies the more pronounced the difference. For logic level circuits a volt or two of noise can produce unwanted effects.


Two opto isolated systems must have separate power supplies that do not share the same ground (common). The whole point of the isolation is have the "galvanically" isolated.....no chance for current to flow for any reason.

Since the Campbell card comes with two separate supplies on the card and they are already hooked to the right place all that issue is resolved.

I am not a big fan of triac based (solid state) relays unless you KNOW what the load is going to be. Highly inductive loads can cause triac misfires and jittery operation. That noise can be reflected back into the AC line. They don't work with softstart router controls or DC anything...including logic to VFD's or a plasma torch on signal. Conventional electromechanical relays are more reliable in a world where you never know what will be hooked to them.

Speed controls made for routers use the fact that a universal motor's RPM is proportional to the applied voltage....unfortunately so is the torque. The same is true for DC motors. At a given voltage when you load them the RPM drops. Unless you have some form of torque feedback from the motor maintaining anything close to a constant speed with a simple phase controller is impossible. As you crank down the RPM it's more pronounced. Constant RPM with an AC speed control is fairly straighforward. For a DC motor it's more difficult.

Sometimes the devil is in the details. Just adding components to a board does not always solve the problem....than takes engineering.

Tom Caudle
www.CandCNC.com

I'm using the Campbell breakout board plus not the combo board and there is no onboard spindle control. I know the cnc4pc board has spindle control and if memory serves it has a lot of features for half the price of the combo board. I don't know about the combo board but home/limit is a pain with the breakout board plus. I originally went with the Campbell board because it supplies the 5volts for the geckos.
I'm sure everyone's experience is different but I was never able to get ANY customer service after numerous emails to Campbell and they have no phone number. Anyway, I'm planning the cnc conversion of my lathe and will be using the cnc4pc board. They have answered my email questions within 24 hours and they run at my friend's house 8-10 hours a day.

I finally had some spare time to wire up my C11G breakout board to my Gecko 203v drivers. Everything (so far anyway) is working as well as I had hoped. No glitches and my motors are running much smoother than they did with the outdated Centroid boards. I spent an hour or two with a ballpoint pen in my spindle and a piece of paper taped on a scrap aluminum plate drawing shapes as I ran through some of the wizards in Mach3. I finally felt confident enough to put an endmill in and cut a 2.5" circular pocket which turned out to be a 2.321" pocket so I have some step count tuning to do but I'll work on that in the next few days. I only ran the X & Y since I need to do some work on my spindle but hopefully in the next few days I can get it back togther.

Anyway- back on topic, I don't have any exposure with the Cambel board, courtesy of this threads suggestions, but from my early experience with the CNC4PC C11G board, I feel safe to recomend it. It was shipped quick and my emails regarding the sale were answered within a couple of hours. The directions were adequate, but for a novice such as myself, a little more detail would have been nice however I managed to get it all togther without asking for service support.

i just purches a breakout board from cambel designs. i removed the cheapie board and wired in the new board. i have only the 110 volt and the step/dir/com wires to my gecko 203v 's. when i try to jog the motors will no reverse. it appears my pin assignments are correct. running mach 3. with my meter i am not geting any direction switching on the dir terminal out of the board. help.

nevermind i got it working

nevermind my nevermind, i only got thedrives working by changing pin asignments to x dir2, x step 3, y dir 4 x step 5, z dir 6 z step 7. bob cambel said this should not work. none of myhome switches are working with cambels recomended pin asignments. im back to me cheapie board.
i would like to get my cambel board working. nothing works right when i use thier recomended pin asignments. i tried two boards, two computers and same results. any ideas?

The proper pins are X Step 2, X Dir 3, Y Step 4 Y Dir 5 etc. You have to use the correct polarity to the Gecko Common (PC Ground) AND the Step signals HAVE to be POS (red X) in the Ports & Pins/Motor Outputs. Probably the reason it's working like you set the pins is that the polarity of the signal is correct (Dir polarity just determines the direction the motor spins. Go back and set the pins correctly and flip the Motor Output STEP active low on each axis to POS (red X).

TOM CAUDLE
www.Candcnc.com