Vacuum pump location?

[mawa]

Currently building the machine I started planning the connection between the gantry and the rest of the world.

I intend to keep the single cable count as low as possible by e.g. using 12 core chainflex cable which runs perfectly since years on my CNC mill.

An eyesore for me is the routing of the long tube from the pump at the front of the machine ??? to the solenoid and its power wire from the FET switch/ tinyG.

Therefore the question: Is there any reason why the pump is not directly mounted to the gantry and is mounted at the right front?

• the pump is not heavy and the additional power wire is easy to add.
  The tube length would be only a couple of cm.
  Maybe the solenoid could even be omitted, because you probably only activate the pump when the task is to transport a part.
  Finally, if the pump can reverse its direction, you could even think of activating it via a reverse bridge circuit and PWM to control the pressure. Then you could let the placer "spit" the part.

The later idea comes from a problem I encountered with my manual P&P device where the vacuum degrades slowly and some parts "stick".

Or am I thinking wrong?

[mrandt]

Hi Manfred,

I guess the vibration of the pump could cause accuracy issues if mounted to the Gantry. Maybe that would also increase wear and tear on the machine.

To eliminate vibration and also to reduce noise, I did not even mount my pump to the table but put it into an insulated box which is placed on the floor next to the machine.

I agree, long tubes cause issues - the longer the tubing, the longer it takes to evacuate the air in it and vice versa. You could almost think of it as a capacitance problem

I think the solenoid valve mounted close to the needle and timing in software (delays after pickup and place) help overcome most issues.

Another approach - which I followed - is to use reservoir tanks for negative and possitive pressure and a solenoid valve for each connected to the needle via a T-piece. This implements what you suggested - "spit" the part onto the PCB by releasing a small "puff" of compressed air into the tube. Thus, I no longer use a 2/3 solenoid on the gantry and am able to release parts almost without delay. But it is hard to judge wether that's worth the effort.

Regards
Malte

[JuKu]

> Is there any reason why the pump is not directly mounted to the gantry and is mounted at the right front?

Not really. reducing the weight adds speed, but you are right, the pump is not heavy.

> To eliminate vibration and also to reduce noise, I did not even mount my pump to the table but put it into an insulated box which is placed on the floor next to the machine.

But this makes sense.

I haven't tried how much different pump locations would affect timing. My gut feeling is that the long tube from the valve to pump helps, as there is some vacuum built up already. Also, it hasn't occurred to me to try running the pump in reverse. I haven't found the need for release puff, it seems to me that the valve produces some anyway; at least the system has released all parts for me just fine so far.