Vacuum setup

[bobgee...@gmail.com]

Hey everyone.

I am just about to go and fetch my nozzle holders from the postage depot. These are the last parts I need (I hope) so am about to start the wiring and software setup.

Huge thanks to Matt and Daniel for making me some of the custom parts, 100% couldn't have got where I am without you guys.

I have fairly closely followed the BOM for Anthony's design and have built everything.

I have some questions about how to setup the vacuum.

I purchased two of the robot digg solenoids. I can't find any info on how to setup the vac apart from this link:

https://github.com/openpnp/openpnp-openbuilds/wiki/Build-Instructions

They are using 4 solenoids and not the same as mine:

I have 2 of these:

http://www.robotdigg.com/product/566/High-frequency-Solenoid-Valve-24VDC

With 4 I am supposed to connect like this:

How do I connect 2? Can I use 2?

Also, from the same link on the github, they say you should attach flyback diodes.

My solenoids have wires coming out of them and go straight to the smoothieboard, can I put the diodes accroos teh back of the terminal block where I attach the solenoids to the smoothieboard?

Do I need the flyback diodes?

Any help much appreciated!

[Reiner Schmidt]

They are probably best attached directly to the solenoid. Then they are there for good and wont get in the way and make the smoothie board messy. That being said its not going to hurt anything if you attach them to the terminal blocks.

[BZHEX]

Your solenoids are different from the ones used in schematic, because they have 3 ports. Put vacuum at port 2 and you can switch between 2-1 or 2-3 for air flow. Normal position is 2-3 ( not switched ) so you leave port 3 open for air access. On port 1 connect nozzle, and when you switch the solenoid, air will access through nozzle creating suction. So you need 1 solenoid for 1 nozzle. I used 2 solenoids because i need to "blow" a part of the air to push the components away when not picking, and for that need a 2nd one.
Diodes are good to use whenever you have inductive loads ( coils, motors etc ) on MosFETs. Some MosFET have internal diodes, but it is safer to use external. Yes, you can connect them to terminal block.

[evilwulfie]

On 1/28/2017 1:55 AM, bobgee...@gmail.com wrote:

Hey everyone.

I am just about to go and fetch my nozzle holders from the postage depot. These are the last parts I need (I hope) so am about to start the wiring and software setup.

Huge thanks to Matt and Daniel for making me some of the custom parts, 100% couldn't have got where I am without you guys.

I have fairly closely followed the BOM for Anthony's design and have built everything.

I have some questions about how to setup the vacuum.

I purchased two of the robot digg solenoids. I can't find any info on how to setup the vac apart from this link:

https://github.com/openpnp/openpnp-openbuilds/wiki/Build-Instructions

They are using 4 solenoids and not the same as mine:

I have 2 of these:

http://www.robotdigg.com/product/566/High-frequency-Solenoid-Valve-24VDC

With 4 I am supposed to connect like this:

How do I connect 2? Can I use 2?

the valves have a vacuum port, an exhaust port and an input port

exhaust port has a little screen and goes to open air, vacuum port goes to pump and input goes to the head.

Also, from the same link on the github, they say you should attach flyback diodes.

My solenoids have wires coming out of them and go straight to the smoothieboard, can I put the diodes accroos teh back of the terminal block where I attach the solenoids to the smoothieboard?

Do I need the flyback diodes?

Yes i added them in line to the wires

Any help much appreciated!

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/063eb8a6-b645-4ccf-964d-9e38eece663e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Rich Obermeyer]

For those that read "some MOSFETs have diodes built in".  Forget it.

That is a body diode and will do nothing to protect the MOSFET from

the kick back voltage from the inductive load from destroying the MOSFET.

An external diode is always required.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/4814afc6-19c0-5ecd-73b3-eb9c6605f2fb%40gmail.com.

[bobgee...@gmail.com]

Thanks to all for your help so far. Makes much more sense now :-)

Just realised I don't have any 1N4001 diodes lying around. I do have some 1N4004 and 1N1587 diodes.

I guess the 1N4004 are the closest match and would work fine, but does anyone reading this have a strong opinion on how to proceed?

Cheers

[evilwulfie]

4004 is fine

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/31a84471-0c62-4ede-8620-5672dc779569%40googlegroups.com.

[bobgee...@gmail.com]

Cheers for that, I figured would be fine, but always nice to get a second opinion, don't wanna blow the most expensive part of my build based entirely on my own bad judgement!

[Michael Anton]

Actually, this is not correct.  There are MOSFETs with avalanche rated diodes that can be used for flyback protection, such as the IRFZ44 for instance.

[Rich Obermeyer]

So Michael,
Which PNP controller has that MOSFETs in them?

> On Jan 28, 2017, at 5:46 PM, Michael Anton <3d.m...@gmail.com> wrote:
>
> IRFZ44

[SMdude]

I put general purpose diodes on mine up at the head where the solenoids connect into the main loom. This way it keeps the flyback current as local as possible, minimising interference, but it probably doesn't matter! :D Old habits die hard!

I really need to think about doing a head mounted controller, there are too many wires!

[bobgee...@gmail.com]

Final question on this (hopefully):

The solenoid from robotdigg has two grey wires and no obvious polarity marked on it. Does it matter which it is wired, if so, how do I tell?

[SMdude]

Nope, doesn't matter. Just make sure you put the diode the right way though..

Final question?? !! Haha! :D ;) There's always another one!

[bobgee...@gmail.com]

Thanks a bunch, that's what I expected but good to get a second opinion.

Not really about the vacuum, but another question about wiring springs to mind. Technically, that's not another question on this topic, so I am still on track!

I guess most people are using the robotdigg Nema-8 motors with yellow, green, red and black wires? Could someone please tell me which way to wire these to the smoothieboard motor connectors? Am I right you just need to put the pairs for each coil side by side, but in no particular order? If so, which are paired? Usually green and red are a pair right? My other motors have a blue, not a black is the only reason I am not sure.

Thanks

[SMdude]

Check here: http://smoothieware.org/3d-printer-guide

[bobgee...@gmail.com]

Cheers, that helps.

So they show blue, yellow, green then red. For the robot digg motor, do I jsut swap blue for black or are the other colours different also?

Thanks

[Michael Anton]

I never said there was a PnP controller that used one, but you seemed to imply that they didn't exist, so I pointed out one that did (and an old one at that).  I'm sure there are many others, so perhaps the odd PnP controller actually uses one, or has appropriate protection to handly inductive loads.  As you have stated, without checking out the design first, it is not a good idea to assume that the board can handle inductive loads, but stating that diodes must always be put in is also an assumption, even if it is a safer one.  They may not be required, depending on how the output was designed.

Also note that a flyback diode is not always the best way to handle this if you want the solenoid to act quickly.  The standard flyback diode keeps the current recirculating in the coil longer than if one allowed the voltage to flyback by limiting the peak with a zener.

[SMdude]

Best to check with a multimeter and wire them as per the diagram.

[Oz-Ron]

Michael & Maddog,

 

You are both correct and both wrong at the same time. Deuce!

 

Can we please keep Jasons OpenPnP forum civil (and especially friendly).

Jason works his ring off on this project (and supporting it) so I am sure he doesn’t need this kind of distraction seeping in.

 

Please continue to revert to constructive, informative & fun stuff in the name of Team OpenPnP!

 

Thank you,

Ron

[Rich Obermeyer]

Thank you Ron for being wrong too.

My originally point if you read it was that using the internal body diode was not a good idea without a bunch of technical discussion. As I said "forget it. Always put an external one. " No harm to having two in this application.  Don't see how that was wrong.  

For those that are asking whether they should put an external diode, I would say that was good advice and was mimicked by Michaels response too.  Obviously those asking the question did not, don't want to or do not know how to figure out the technical details of the responsible MOSFET.  Just looking for a quick answer.

Interesting that the discussion of what voltage and current the diode should have was not pointed out.  Just "use any diode lying around".

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/3d11977a-ddfd-4cfd-9c41-979a294ffc7f%40googlegroups.com.

[Michael Anton]

I don't see anything in our discussion that was not civil, and friendly.  Consider it additional technical information (it was constructive), which should always be welcome.

[Michael Anton]

Just another FYI, since the original picture provided looks like a Smoothie board, I looked up the schematic.  The large MOSFETs used are avalanche rated, so they will be fine for inductive loads as is.  The small MOSFETs, have appropriate flyback diodes in place, so they will be fine too.  So, on this particular board at least, no external diodes are required.

[Oz-Ron]

OK, very glad to be proven wrong - once again.  It is fantastic that this forum is friendly and wealth of helpful information shared.

 

Back to the subject, and my 2 cents:

It is good practise to put the back EMF suppression diode close to the inductor (in this case the valve) as it helps reduce radiated EMI. On equipment like our PnP machines EMI can be a source of unexplained malfunction so anything that minimises the interference is worth incorporating.  (If you rely only on the Mosfet body Avalanche / Zener diode without any diode to the positive rail at the coil then the energy in the coil is radiated Marconi style on the way back to the switching device.)  Many automotive relays have diodes & resistors incorporated for this reason.

 

The type of diode *could* be almost any diode, except if using a signal diode then it is best to add a low ohm resistor (~10R) in series to limit the current.  Another consideration is when clamping a PWM controlled load back EMF, the switching frequency may dictate the need for a fast recovery diode otherwise the diode will dissipate excessive current.  For a straight on/off situation like the pick valve, I agree, in my experience a 1N4001 or 1N4004 is all you need.

 

Cheers,

Ron

[Michael Anton]

I recommend you try a zener (actually a zener in series with a diode), to see if that decreases the valve turn off time.  To get the valve to turn off, you need to get the energy back out of the core, and the fastest way to do that is to let the voltage fly to a higher voltage than just a diode drop.  So, try replacing the flyback diode with a diode in the same configuration (cathode to the supply), but on the anode end, connect the anode of a zener, and the cathode of the zener goes back to the coil.  Putting both diodes still helps keep the currents local to the coil.  If you are still worried about EMF, you could also put an RC snubber across the coil to catch any fast spikes.

If the coil was driven by a PWM signal, then using a zener diode would be less desireable, as you would want to keep the current in the coil, rather than dissipating it.

[Michael Anton]

On another valve related note, driving the valve from a higher than rated voltage supply (say 24V on a 12V valve), with a series resistor to limit the current to the normal valve rating, would be a way to get it to turn on faster.  The penalty is power dissipation in the resistor...

On Tuesday, January 31, 2017 at 4:12:38 AM UTC-7, Oz-Ron wrote: