One of my daughters is dating an electrical engineering major so I'll see if he can come up with a circuit board printout I can etch and drill. Mr. Richards, do you have a source from which you would recommend I order these parts?

Mr. Richards,

Is it possible that the part number on that proxy is BALLUFF ES-M12EL-NSC40B-S04G rather than SG4G? I couldn't find SG4G sensors.

Doug,
I just cross-checked the Balluff catalog (sensor catalog '05). The NPN sensor that I use is found on page 1.32 and has the part number BES M12MI-NSC40B-. The suffix, S04G is only found on the plastic parts bag. That suffix might be the quick-connector number that mates with this particular sensor. I bought two more sensors and some more cables from Fiero Fluid Power. The sensors cost $55 each and the cables are about $14 each, depending on the length. Today they only had 7.5m cables in stock.

I'm sending a schematic with board layout to Gerald. Everything was done using ExpressPCB, which is a company that provides quick turnaround circuit board manufacturing. They also give away their software to make designing simple circuit boards fairly easy. Their fastest prototype service costs about $60, which includes three 2-sided boards (without silk-screen or solder-mask). The only drawback is the board size for that service, 3.8 X 2.5 inches. This particular design just barely fit that layout. That same board with silk-screen and solder-mask costs about $11.50 each when thirty boards are ordered.

This PC board kept me awake last night.....it needs to be thought through a bit more:

- Location of board. To minimise wiring/cabling, this board should be riding the gantry or the y-car. Proxy's typically have 2 meter [6ft] leads - if all 4 proxy's have to reach the board without extensions then the location could be immediately after the y-chain.

- Housing. Since the board will not be in the control box, a housing needs to be selected and the board sized accordingly.

- Gantry autosquare. The 4th proxy for the slaved gantry motor needs to go independently to the BOB if autosquaring is desired. Maybe a jumper option to have 4th proxy in series or independent?

- Z-zero. Many folk will be using a touch plate to zero the cutter height to the workpiece or the table. This can be in the same series circuit as the other switches, but the logic is reversed (contact is made instead of broken). Can terminals and chips be put on this board to connect the z-zero wires and reverse the logic?

- Junction central. Since we have already established this board is riding the car, in a housing, gathering all the proxy leads together (+z-zero) and has cabling to the PMDX, it might be useful that this board be expanded with more terminals to break out to all the pushbuttons (low-voltage) on the gantry. In effect, have this board extend the BOB all the way to the gantry . . . . .

Food for thought

Mike, Thanks for pointing me to the source and for designing this circuit and board layout.

Gerald, I don't know what I was thinking when I wrote my post last night. I can put together kits containing loose parts and ship to MM builders for my cost. With all of the help I've received during my build, this is my chance to pay something back.

I was thinking about the PC board too. It occurred to me that I might be able to use the MM to mill the cladding off if the components aren't crammed on there too closely. At least for the prototype anyway.

Gerald,
Normally, I build everything around a little 20-pin microcontroller so that features can be added with software - instead of a hardware redesign. The limitation of using a microcontroller is that most of them require a special programmer to write the program into the chip. The programmer that I use is fairly basic in operation, but it still costs about $1,000. It's had plenty of use on lot's of projects, but it isn't something that everybody would want to have on hand.

When this discussion first started, I thought about building a 'proper' auxiliary controller with 22mm bypass switches for each proximity sensor, 22mm indicator lamps to show the status of each sensor and as many other 'features' as the 20-pin microcontroller would allow, but just the 22mm parts and pieces for that device would cost well over $100.

However, if you can come up with a useful design that can be implemented with a microcontroller, I'll be happy to program chips for anyone who doesn't have access to a programmingr device.

Mike and Gerald,
I have access to a programming device as well if anyone needs devices programmed. Or even for board building, for the soldering challenged. :)

For those that havn't noticed it yet, more has been added to Mike's post with the schematic (post #37)

Mike, the programmed microcontroller approach is probably the most elegant. It would put those of us in Africa, India, etc. out of the loop for "upgrades" which are likely to be numerous in the beginning. But I have enough patience to hold back on getting the first few versions.

My approach to these things is from a hardware perspective, as you might have noticed above; It must fit in a box that rides the gantry. Leading into that box:
1 cable via the chains to the PMDX inside the control box
4 cables out to proxies
3 cables (one each) to the three pushbutton stations
1 pair of connection points for a touch plate (normally open)
1 pair of connection points for a touch probe (normally closed) ^{*new requirement}

Relays for limit switches
 

A little something I found at Winford.com. An electronics supplier I use often for small break out relay boards and isolation contacts.


The board is available with three different relay coil voltages (5, 12, 24 VDC), allowing you to choose whichever is most convenient for your project. Please note that the relay coil voltage you select has no effect whatsoever on the voltages you can switch with the relay contacts.
For mounting on a panel, we suggest the rubber feet version (-FT), 1/4" nylon standoffs under the mounting holes, and #6 screws.
For more information, please see:

• http://www.winford.com/img/pdf.gif RLP104 Datasheet
• http://www.winford.com/img/pdf.gif RLP104 Schematic

Features

• LED indicator for each relay
• Relay contacts rated for 15A at 250VAC
• Screw terminals for relay coils (accepts 16-26 AWG wire)
• Screw terminals for contacts (accepts 12-24 AWG wire)
• Approximate dimensions: 2.6" x 5.9"
• Rubber feet or DIN rail mount

DIN Rail Mount Version

These products are also available with DIN rail mounting clips for quick and convenient mounting where DIN rail is used. Product numbers ending with -DIN come with DIN clips already installed and ready to mount. The clips allow mounting on standard 35mm rail as well as 32mm rail. http://www.winford.com/products/pic/rlp104-12v-din.jpg
(RLP104-12V-DIN pictured - rail not included)

Sean for that application I prefer to use this style:

Gerald,
Funny you looked that one up! I use them regularly from FactoryMation. Exceptionally small form factor and they have both NC and NO contacts. Plus, 3 voltages to choose from 12vdc, 24v ac/dc, 110v ac/dc.

The PC board in the previous post was for discussion/reference only for those interested in making a board of their own. Not for me...I like machine automation gear - easier to replace.

Bus bars are available for these relays.

If you look at the accessories page associated with this relay, you will see they sell them in 20 & 30 connection lengths...cut to fit. Three colors available.

Gerald,

What you've described should be fairly simple to construct, but, we'll need some more information.

PMDX:
How many signals to the PMDX and what polarity (active High or active Low)? (I'm assuming three signals; one from the proximity switches and two from the touch plate - one active high and one active low.)

Proximity switches:
NPN or PNP?
Normally Open or Normally Closed? (I use NPN, Normally Open)

Pushbutton Stations:
Are all three pushbutton stations linked serially?
What do you want this auxiliary box to do when a pushbutton on one of the pushbutton stations is pressed?

Touch Plate:
Normally open contact: Is this contact Active Low (input always sees +5V until the cutter touches the plate)?
Normally closed contact: Is this contact Active High (input is at ground potential until the cutter touches the plate)?

Voltages:
+5V
+12V
Common Ground
(Voltages supplied from the control box via the PMDX cable?)

L.E.D.s:
+5V present
+12V present
One L.E.D. for each proximity switch?
One L.E.D. for touch plate N/O
One L.E.D. for touch plate N/C

Bypass Switches:
Do you want switches to bypass each proximity switch?

Mike, I am not ignoring you, just giving all of this careful thought. . . . . . .

Would love some other folk to chip in with wish lists as well.

In my mind's eye I see it as a way of bringing the PMDX-122 (at least a large part of it) all the way to the gantry. The part of the PMDX that stays in the control box "drives" (geckos and relays for dust collectors, etc) - the part out on the gantry "receives" (limits, homes, p/buttons, touch plate, probe)

Out on the gantry we have lots of wires to be terminated and held securely.....I see mostly a lot of stupid terminals and relatively little board space with smart chips on it .......

Anybody else?

As long as we get proximity switches, I'll be a happy camper. I'd love to add the laser pointer but I don't know that we necessarily need it to be connected to the PMDX. Might be nice to turn on and off with a pendant so I guess a connection to the PMDX would help in that regard.

I posted a while back but cant remember which thread about a company on ebay that was selling boards that worked in conjunction with the bob to enable connections for limit switches and e stop functions but nobody ever weighed in with their opinions on it.

Craig

Here it is:

I'm not sure if this is the right place for this or not but I have been watching a sellers store on ebay, http://stores.ebay.com/HUBBARD-CNC-INC he has limit switches for sale cheap but the things I really wanted to were the power supplies he has and he has (for lack of better terminology on my part) a pass thru db25 connector that is supposed to protect your computer and it breaks out the e stop and limit switches all in one. It has USB connectivity on one of the models. If I understand it right it goes between the pmdx and the computer so does it give the ability to ouput mach to the controller via usb and does it make it easier to set up e stops and limit/homing switches?

again this loosly fits this thread but feel free to move it and if anybody with more knowledge about this would look and give their input I would apprieciate it.

You just needed to do a search on your name.

yes thankyou, thats the post, he however has other boards besides that one that I was refering to that handle limits and estops maybe something to look into but as I am the least informed I just wanted to point a direction.

You'll have to forgive me if I'm barking up the wrong tree. :)

With the simple touch plate that I advocate, the second option is not so desirable.

It implies to me that the tool is at 5v. This may come into conflict with normal machine grounding.

Greg


EDIT: I see on more careful reading that this was not what Gerald originally suggested

"1 pair of connection points for a touch plate (normally open)"
"1 pair of connection points for a touch probe (normally closed)"

Probe not Plate. :)

Here is my proposed proximity homing/limit switch layout w/ relays.
Prox switches rated at 10-30VDC...I am supplying 12Vdc from output on Antek transformer. I ordered 5v, 12v and 56v outputs on my transformer.

Based on the wiring logic supplied by factorymation pdf...I believe I have this correct. Any takers to verify my schematic?

Sean

Attachment 987

Your schematic shows +12 and -12V, therefore 24V relay coils. Was that your intention?

12v coils.
I wasn't infering a differential voltage of 24 volts...only 12volts. I should correct the schematic to reflect otherwise.
The system, ideally, would be using the 12v I have native in the box and not introduce another voltage.

Thanks for looking at this.

SEan

My only other "schematic" comment is that you indicated one proxy each on each axis. My choice would be definitely 2 proxies on x-axis, 1 on y-axis and a "maybe" for the z-axis. The z-axis hassles are when it dives too deep and a fixed proxy can never detect that. (Some folk have sliding limit switches on the z).

Sean,
Those proximity switches must be Normally Closed for your circuit to work, otherwise you'll only get a 'contact' signal when all three sensors fault at the same time.

Mike the metal targets will be present in front of the proxys in the "normal" condition. In the presence of that metal, the NO switches will close.

The targets are "abnormal" in this case - we are not trying to sense bolt heads or lumps of metal at the ends of travel. We look at the rail all the time, but there are holes at the end of the rails. When the proxy sees the hole, it changes state and opens the relay. Also when the car jumps off the rail.

Gerald,
Sorry. I should have remembered that the sensors are actually going 'active' when they sense the 'hole', which, as you pointed out, makes Sean's circuit work perfectly.

Gerald,
Thanks for verifying the "logic". I assumed we wanted to see the rail at all times incase of a "jump the track". Thus NO logic on the Prox.

Glad I ordered the correct thing....no canceling that order now :)

Are these proximity switches or sensors necessary? Is there an advantage to have them or are they just a nicety? I see three of them in some of the posts, why not six?

What is their purpose?

Can I do without these proximity or limit switches to start out with and add them later?

Hey Ed,

Gerald wrote earlier that he still doesn't have switches on his machines. I don't have them on mine at the moment but I think I'd like to have them. I guess I'm just clumsier than Sean D. So I guess the answer to your question is that they're a nicety.

You actually only need four switches or sensors. One on both ends of the gantry, one on the car and one on the Z axis.

You can definitely start out without them and than add them later.

Ed,

Agree with Doug. You certainly don't need them to get going or to have a productive machine.

I'm adding the prox. sensors, shuttle Pro, and Z-zero plate to make the mundane, every project essential starting point/process more efficient. I don't want to spend my time zeroing or finding my hard/soft limits for each new project.

You can add all or none of these items at any point. That is the beauty of building your own machine. Like any good receipt, you start with a good base (Gerald's design) and "add to taste".

Ed, (and others)
It took me less than 6 hours to mount,run all the cable, terminate and configure Mach 3 to work with the proximities. Gerald did all the hard work - he gave us a place to mount the sensors and a way for them to activate at the end of each axis. The only axis not accounted for was the z, which I made the mount, tapped the holes and painted in the above mentioned time frame. Cost was about 150.00 with all the cable and shipping. DEFINITELY a time saver when setting up the machine. Plus, the work offsets are nice to come back to after a "morning" home. I just click "ref all", click offset....and were cutting. ....almost shop labor proof...."almost" :)

Summary of the Proximity Switch usage
 

I have fielded a number of Emails and PM's recently to explain my installation of the proximity sensors on the MM per Gerald's design intent. A quick summary below based on my choices and design input from Mr. Richards, Doug, Gerald and others.

MechMate Proximity Limits

Schematic for installation: (this is correct, NO contacts wired in series to the PMDX input)
prox_layout.pdf

Details: FactoryMation
qty description
3 SI12-C2 NPN NO H m12QD proximity sensors, 10-30vdc
1 *cant find the part number* prox sensor like above with 30mm body length
4 PIR6W-1P-12VDC slim relay
3 MOD.14/4 LC5 5m qd cable. (right angle connector) - x and z axis. I extended each cable with 4 core cable to meet my needs.
1 MOD.14/4 LC5 5m qd cable. (straight connector - ycar only)

Please note that the prox switch is NO logic
The relays are wired NO logic.

Sean,

You forgot that you actually need 4 sensors. 2 for x and 1 for each y and z.

The diagram need one additional sensor added. I only know this because Sean and I talked about it.

Nils,
Correct, I didn't update the schematic like the parts list I just posted.
All still applies. Just add 1 more sensor in series for the x axis for a total of 4. Good catch.
Thanks,
Sean

In my mind's eye we still have a "box" riding on the gantry, with a logic board inside, as discussed in this thread a month ago. The blue thing in the sketch below. Realise that there are 35 wires and 4 screens to be taken care of inside there . . . .

Gerald,

What you've posted is a classic multiplexer/demultiplexer, which is just another name for 'switch box'. The next thing that we'll need is a chart that lists each device and its destination, such as:
Push Button Station 1, Red PB (N/O) -> Pin 11
Push Button Station 2, Red PB (N/C) -> Pin 11

From that, a circuit can be drawn and a circuit board can be built.

I just get drawn into the forum whenever Mike post's something.
It's like watching Einstein discuss the theory of relativity. :)

Ed,

Gerald's post answers the NO question.

I'll try my hand at this but I may need help with the Z zero plate and the black button on Gerald's 4 button station.

Sean, or anyone

Ed asked myself a question that I can't answer. I thought I understood the wiring and logic, but I don't.

On the relays, what pin numbers are you using for the latching side? In your earlier photo, it looks like pins 12 and 14. That makes no sense because its an open circuit :confused:

Pins 11 and 12 makes sense, but that would be NC and you said it's NO. Also, on three NO relays in series, how can the BOB recognize anything when only one out of three switch's closes.

A1 and A2 are jumper-ed and that makes sense.

Greg,
Pins 11 and 14 of the relay are wired in series to the input #15 on my PMDX.
It's open when not energized, closed when energized. Whenever the machine is powered "on", it's energized and closed.
When homing or seeking limits, the proximity switch "sees" the hole in the rail, opens the relay, opens the switch and activates the pmdx. Then, because the Mach routine for homing is smart. Mach will seek the switch, activate, them reverse the axis until it is cleared then move to the next axis. Mach's sequence is designed around using multiple switches (regardless of type) to utilize only 1 input on the BOB, thus freeing up inputs for other things. Like a pause and resume button on the gantry, or z-zero function button.

Page 8 of the pmdx manual show's "loosely" multiple switches configured to a single input for this reason.

Ed and I have exchange multiple emails over the past number of days discussing the NO / NC reasoning.

I like the switches powered in parallel, but input wired to the BOB in series via a NO contact. I have many reasons for this, but mostly a NO contact would show a failure to the BOB if one of the sensors is not powered on or a wire is broken because the switch MUST activate and pull in the relay to work. A NC switch in this case may not show a fault until you try to home or hit the limit, then you would find the machine crash into the hard stop....or you get to listen to the z-axis grind teeth for lunch.

Let me know if I have explained this logic well enough...I feel like the NO / NC switch choice battle isn't over on the forum yet.

For me, my MM works like a champ set up like this and I have even de-railed the machine on purpose to test me theory....its works well.

*Gerald and Mike can chime in here and help me any time now :)*
Sean