The fridge keeps the shims fresh.

Cable Question for the Experienced Forum Members
 

Hi All

I have been doing more on my build but just nothing really worth taking a photo of, so for the moment that side of things is slumbering.

I have a question re the suitability of a cable Skippy has bought for us.
It is a Belden 6+E .5 mm multi stand conductor cable that has a 10.7 mm diameter and 85% tinned copper shielding. See the PDF Spec Sheet attached for more detail.
The cable chain this cable will be routed through has a 70mm bend radius, the closest we could get to the 75mm plan specification for the bend radius.

What I would like to ask is, if this cable is suitable, as using the rules of thumb it exceeds the bend radius by a good measure.
The availability of the "perfect single cable" that has lots of cores AND a tight bend radius is basically zero here in Australia.

So what have others done about this issue or what experiences have others had putting a similar thickness cable into a working mechmate.

Thanks in Advance

Ross

Cable
 

I am using the LAPP Olflex Classic 115CY 4G1 Part No. 1136204 (4 Core + Shield)

This cable will be used for everything both internal and external to the control box and will even be used to run my Makita Router @ 2200W running 240V through it by paralleling the 4x strands @ 1.0mm each to make 2x 2.00mm for both Active / Neutral.

My theory when choosing this cable was that i was less focused on getting heaps of strands into a single cable and used the theory that if i needed more strands i just ran another cable from the control box.

Thanks for your input Mark, some good sense there.

Regards
Ross

Marking Up The Spider
 

My Spider came back from the laser cutters less the all important alignment marks for the six V bearing bolts.
Marking it up was proving difficult then I came across someone on the forum who had used a permanent marker refill to create a paint surface that can then be scratched to give a highly visible line.
Looked for the post again but could not find it, none the less thanks as it helps us not so familiar with the tricks of the metalworking trade.

Here is a pic of the result.
Attachment 9832

While I'm at it -

Go Kobus with your new UIRobot controllers - I have read all the specs and now even wishing that Mach was CAN aware !!

Regards
Ross

Hi Ross, being cutting all afternoon as we have a holiday today. Those drivers are working like it should....no problems so far. Still waiting for some cable to put the driver ON the stepper to test.

Touch Screen Monitors
 

Ebay Australia has some touch screen monitors (only the monitor panel not the whole internet kiosk) the same as Aus Maddog and Skippy up for auction.
They are set at pretty good price for a used 15 inch touch panel, but you will need to make your own judgement on that side of things.

Check out http://cgi.ebay.com.au/15-Touchscree...#ht_500wt_1154 if you think you might be able to use one in your build.

Regards
Ross

Hello Ross,

As an industrial electrician i can offer the following advice for your cable,

yes you have exceeded the bend radius by enough to cause problems.
the first issues you will experience is broken cores in the cable this could cause problems if u use this for stepper morors depending on driver selection (open circuits blow the output stages).
at best i might guess this cable might last 1-2 years tops. you would be mutch better off using 2 or 4 core cables.

A more expensve option is using rubber insulated cables (may need to made to order)$$$
Regards Andrew

Thanks Duckie

The cable was to be used for signaling and as such the individual cores are only .5 mm2, far too small for a stepper.
I take your advice on the bend radius and a picture of what is the way to go is now forming solidly in front of me as a result of inputs from a wide range of sources.

I have had a significant rethink on cabling and to be honest even my own fall back position on cable for my steppers is underdone.
I might add this is the second time I have bought 50m drums of cable I probably won't use, so it starting to hurt a little.

The real issue here was a supplier of high quality cable was slow to rise to the top of a somewhat limited pool of automation cable resellers.
A combination of rigorous attention to the specifications of the cables, a different wiring topology and a recognition of the compromises I am prepared to accept all form part of the way forward on the cabling issue.

Once again thanks for your free professional advice, hard to get these days !!

Regards
Ross

Proximity Sensors
 

Today I had some proximity sensors arrive from China.

The specifications are ... NPN NO type - three wire - 4mm sensing distance - 12mm diameter sensor.

Attached are 4 images showing the two states of the sensor in operation and the Mach3 screens.

Sensing Mmetal - the "Normally Open" sensor is closed and the LED in the sensor glows red to indicate it is closed.
Attachment 9841
Attachment 9842

Not Sensing Metal - the "Normally Open" sensor is now open, the LED is not lit and the Mach3 screen indicates a limit has been reached and an Estop condition is triggered.
Attachment 9843
Attachment 9844

I have followed Gerald's advice on the type and configuration of the sensors as it allows you to tell immediately if a sensor is working by just checking if the LED is lit.
Further any departure from metal, like a lift off the rails, also triggers a stop.

Regards

Ross

Ross, while you have a bench test going, would you mind seeing what they do with aluminium as the target?

Sensing Steel Vs Aluminum
 

Hi Gerard

Always keen for an experiment, haha

Sensing distance for steel (20 x 20 square stock) was 4.5 mm.
Sensing distance for Alu (6mm plate) was 2.0mm max.
Voltage to the sensor was 12V DC.

The result suggests that one would need to roughly double the spec of the sensing distance for an Alu target.

Regards
Ross

So, if one rides the switches at 1mm above a set of alu rails, it will work. Thanks for testing. (ride at about 2mm for steel)

Ross,
That is exactly what Nils found out while building his machine. He had to use the extended sensing for the aluminum rails. The standard sense worked, but from memory was a little trigger happy!

Proximity Sensor Wring Question
 

I am happy now that the proximity sensors are working well when wired in series.
The fine detail is still somewhat cloudy for me though.

The proximity sensors have a power source of 12V DC. After checking the third wire to ground I see it returns 12V DC to the jumper 5, pin 11, on the PMDX-122 board.

I have seen various circuit diagrams some with a relay(s) and a resistor.
I was wanting to ask firstly if a relay is necessary and secondly if a resistor should be used to limit the voltage back to the PMDX-122 to 5 Volts.

If any one can help out with a diagram of their solution or an explanation of what is required , it would be appreciated.

Regards
Ross

Ross

This is how i wired my proximity sensors and is all working fine for me.
http://i890.photobucket.com/albums/a...helen/Prox.jpg

And these are my relays
http://i890.photobucket.com/albums/a...len/RelayS.jpg

In the Mach3 Ports & Pins -----> Input Signals you have to set the X++,X--,X Home.....etc your port and pin and leave the Active Low option UNCHECKED in order your proxys to work with this wiring.
That is because your sensors will be ON while travelling along the rail and OFF when above a hole or a distance over 4mm from the rail.

In the Homing Limits section in Mach3 set up your Soft Min values to zero, your Slow Zone values to 3, check the Auto Zero and set up your Speed % to what suits you (mine is at 50%).
Checking the Home Neg option will move the axis on the negative position for homing, so according to your setup you must check this (or not) for your A axis.

Sergio

Hi Sergio

Thanks, nice and clear, so that's 4 relays and no resistors.

One more question though, I assume you have not run 12 separate wires out from the control box to the proxies, rather series connected three wires in the machine.

Appreciate the quick response - I'm still hunting for those flat DIN mount relays both yourself and Gerald are using. For the moment all I can find is a DIN base and plug in type relays here in Australia.

cheers
Ross

Looking at the wiring again - you probably have run 12 wires.

Ross

Ross

What's coming out of my Control Box is the 2 wires (+ , -) from my 12VDC power supply and the 2 wires from the BOB (pin11 & GND) for the proxies to work.

The whole wiring, instalation of the Relays takes place in the junction box on the Y-Car.

http://i802.photobucket.com/albums/y...io-k/Fin17.jpg

Sergio

Ahh, the Jbox is where all the action is happening not the control box !
Thanks again for the information and photos.

Ross

Hi
Ross Farnell or radio spares for the flat relays in Aust.

What is the part number for the prox switch or a spec sheet link?
I expect we can eliminate the four relays altogether with a bit of logic.
I am not up to mine yet but if you give me a link to the specs of what proxy is being used I will test with a modified "wired or" to eliminate the relays.
Regards
David

Hi David

The ebay link is http://cgi.ebay.com.au/ws/eBayISAPI....ht_3499wt_1137 for the NPN NO LM12-33004NA proximity sensors.

Specs copied from the page...
Attachment 9864
Attachment 9865

Cheers for the tip on the Australian source for the relays (nearly figured out what they do now but still baby steps).
I will wait and see what you make of the specs and then see what transpires.
Thanks for the offer to design the circuit or at least have a look at it.

Regards
Ross

Tapped the Z 250mm Z Tube Tonight
 

Progress is steady, so here are some pics.
I feel drawing up the hole centers in CAD and applying them to the tube was a mistake, I should have just used the Z slide as a drilling guide.

Mostly what it did was allow me to check the machining of the Z slide plate as I did not do that part myself.
I scratched though the holes with a scribe and then compared that with the drawn overlays - my conclusion - better than I probably could have achieved myself and certainly faster.

Regards
Ross

Ross

If you're keen on a relay board solution then you might check out Dmitriy's
DIY relay board which is fantastic and easy to build.

Check it here : http://www.mechmate.com/forums/showt...3&postcount=41

And here are the instructions : http://www.mechmate.com/forums/showt...4&postcount=61

Since you're using the same proxies as me (NPN NO) and if you're going for the relay solution then i think there's no other way of wiring them and you cannot omit the relays.

Sergio

Your solution seems less complex for me as a learner in electronics and easier to maintain when something goes wrong.
Those 6mm relay bases are over 20 euro each here now I have found them and that is not counting the cost of the relays.
For some products we really do live on the other side of the planet it seems.

Regards
Ross

Hi Ross
I took the couple of hours to search this site and sure enough:
http://mechmate.com/forums/showthrea...highlight=Prox

This has been suggested before.
In the above thread Mike Richards shows a better digital way to install the prox sensors without relays using optoisolators. ( An excellent elegant solution, I think) As he pointed out earlier ther is also a a way to do this with a single logic gate, which is what I was thinking of.
Using NPN NO sensors is a good choice for an emergency stop as all the sensors (and relays if used) need to have a current flowing through them in their normal operation.
I think this might be nasty when one wire is intermittant after five years of cutting every day ;) as the cause of a "random" stop is not tracable.
I think the relay solution goes from digital to mechanical (relay) back to digital in the BOB. The small "wetting current" through the relay contacts is not ideal and the mechanical relays moving IMHO is not ideal.


I like Mike richards solution with the optoisolators.

For a simpler STOP system rather than treating the limits as emergency stop is to select NPN NC sensors and parrallel them so if a hole is under a single sensor the limit input to BOB is then triggered. As this only has current flowing at the limit it is not "failsafe" and would require periodic testing. It would not stop the machine when a wire breaks.
This would require no relays and proxies would not be isolated from the board.

Anyway it is now late enough to go outside and weld my table together without waking the neighbours.
I will order my proxies next week and am happy to model/test a few sytems if their is enough interest on this site.
(I just dont like seeing the digital to mechanical to digital thing happening even though I know it is "easier to understand" when it is first made.)
Cheers
David

Hi David

I have read all of Mike's stuff and frankly I just do not understand it as it is well outside of my abilities.
If the proximity thread was printed in ink it would be nearly worn though by now as I have read it that many times.
There is a significant assumption of knowledge wound throughout the proximity thread that has made it no so useful for someone of my limited electronics skills.

Sergio

I have modeled your circuit based upon the SM Reish circuit and it will not work for me.
Connecting a prox directly to 12V I get 10V (not sensing) and .8V (sensing) off the black wire.
Given this the relay coil is energised by two positive 12V sources and buzzing of the coil ensues in either state and voltage.

Connecting the coil to side of the relay to a battery charger + and - terminals gives a far less concerting click and all is well.
Also rechecked I am using NPN NO proximities.

Any ideas - last two days has been filled with wires and relays but outside of passing one prox signal though one relay on the secondary circuit I got nutin as they say.

Thanks in Advance

Ross

Hi Ross
I understand it can be confusing.
In the earlier post you showed the sensor working with mach3. What was your wiring?
Can you show a photo of the wiring and any relays you have please.

Do you have spare inputs available or must you use only pin 11?
Cheers

Hi David

Power to prox only test.
+12V on Brown Wire
-12V on Blue Wire
Black Wire is the signal wire

Not Sensing Metal - 9.96V
Attachment 9891

Sensing Metal - .83V
Attachment 9892

Finder 12V Relay Test
+12V to A1 Brown Wire
Black Prox Wire (Signal Wire) Changing to White Wire to A2
Blue Wire to -12V
Coil buzzes due to both Inputs being Positive
Attachment 9893

Proxy Checked to be the NPN NO type
Attachment 9894

I have also used some 30Amp 5 pin ISO type automotive relays with a similar result.
Putting the coil side to +12V and -12V terminal of a battery charger or a 12 volt automotive battery gives a nice clear click.

Hope this is now a bit clearer.

Ross

David - a bit extra for you.

Input 11 is the typical input and all proximity sensors use this one input on the PMDX-122 board. Each proximity switch does not have a discrete input otherwise this would be much simpler but use lots more wires.

The test where you saw Mach3 working was simply the proxy powered up with + and - 12 volt and the black wire going to pin 11 on the PMDX-122. Mach then triggered on the low voltage when sensing. A very simple setup indeed but it worked fine.

Ross

Hi Ross
I suggest that you think of the proxy as a switch (rather than a voltage device. )
The proxy usually gets connected to positive power (yours 12V and ground 0V)
The Black wire in Sergio's post #216 is the ouput of the proxy. as yours is a NO when no metal is near the black lead just floats (That is the Normally Open NO bit.) When metal is detected the "switch is on" and that black wire in #216 is "joined to ground".
If the relay coil is joined to +12V the otherside (-ve) of the coil goes to the black wire of the proxy. The relay should then click when metal is approached.
The proxy is "switching the earth to the relay coil.

(You cant use a large car spotlight relay if it draws too much current.)
Then four of these should be working first.

Switches and relays have a Common a normally closed NC and a Normally Open NO.

It is the common that gets joined to the NC when no power is applied.

When the relay has power (12v connected and the proxy "near metal") then the common is joined to the NO contact.

It is this connection between the COM and NO of each relay in turn (called a series circuit) that means that all the relays have to be "energised" or on for the pin 11 of BOB to be joined to GND of BOB to signal that it is ok to run (ie not at the limit).

When any prox "sees the hole" or moves from the metal it turns is output off shutting off the relay making the COM no longer joined to the NO so a inout is detected at Pin 11 of BOB showing a limit fault.

Hope this helps, you are right the thread on Proxies has a lot of assumed knowledge.
pm me for my phone number in WA if you want to talk through it.
A photo of your wiring, no matter how messy, and the relay will make it easier to talk about.
Cheers
David

Hi Ross

Can you post your relay info and your connections to it ?

David Solves my Problems
 

David has very kindly waded into my electronics woes and has solved my issues very quickly.

In order to leave a remnant in this thread for others to follow, the issue was that I was using a battery charger (out of picture) to power my test mock ups.
David pointed out to me the power the charger puts out is not smooth DC power and it cycles many (50) times per second.
The chatter I was hearing in the relays was a artifact of this poor quality power source. On David's advice I swapped to a car battery and all the problems immediately were resolved.

Sergio - your circuit diagram at post 216 should now work fine with my dirty power source out of the picture.
I will mock up the complete circuit as soon as I can get a full set of the industrial relays and let everyone know how it went.

Currently building two pedestals, one for David and one for Demitriy to stand upon as I'm never going to touch these two blokes in the electronics stakes.
Of course I will hook them both up to a relay with a nice clean power source.

Regards
Ross

Ross

You made the same mistake as i did :D:D

I had also wired my relays to 12VAC power and thus they were buzzing
when i switched to DC everything was fine.
You should have read my adventures first :D:D:D

Keep cranking you're close to dust :D

Sergio

Ross, that relay is also quite big, and its coil may need more current than what the proxy can supply.

Sergio

Yep, should have made the "connection" but honesty I did not have a clue how a 12 volt charger worked, haha. Thanks for your on going monitoring at least now we both know exactly how it feels to look at a pile of relays buzzing when they shouldn't be !!!

Gerald

Thanks as well for the caution on the relay current draw. David mentioned how to check a relay when they had the power across the coil. The sensor I am using has a 200ma current draw so he suggested I limit the coil current draw to 150ma max to provide a bit of overhead for temp and vibration. So much to learn.

Regards
Ross

From post #222, the proxy can drive 200mA.

The Finder 55 series needs 1 Watt to drive its DC coil, and a 62 series could need 3 Watt. Those Wattages at 12V are equivalent to 83mA for 1 Watt and 250mA for 3 Watt.

You can find the Finder specs here: http://www.findernet.com/en/products...rofile.php#ind

Ross, that was a cross-post with your last.

Checked the value for the 56 series relay and it was 90ma so it's a keeper.
Thanks for the link as I was still searching for the specs.
regards
Ross

Did the seller provided an internals schematic of the sensor? I don't quite understand why double power supply rails are needed, as David pointed out it should work with a level that moves near GND and Vdc working like a switch. Thanks!

Pablo, some people say -12V when they mean the negative side of the 12V supply.