Vacuum pump issues

tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Vacuum pump issues

Post by tek »

On the Hello World job, the pump is only picking up the components some of the time. On the last run I did, it failed on the first component then successfully picked everything else up (although the 0603's it picked up at their corners and turned them the wrong way), but then after changing the head, it picked up the IC and then dropped it about half way through moving to the PCB. I've not seen anything about testing or calibrating the pump so is this not usually an issue?
tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Re: Vacuum pump issues

Post by tek »

Further problems include

-machine doesn't want to pick components up from the first reference hole on the tapes, it skips to somewhere on the middle for the 0805's even though on the tape positions table the X location specifies the right position. Though next X keeps increasing by about +30 every time I run a new job.

-machine isn't placing the parts on the correct spot on the PCB. It varies, but seems it can be out by a good 3-4mm.

-just ran another job and when I put the second needle on for the IC it went over the up camera and then cancelled, saying something like "can't see camera".

Cheers
mrandt
Posts: 407
Joined: Mon Apr 27, 2015 10:56 am
Location: Stuttgart, Germany

Re: Vacuum pump issues

Post by mrandt »

Hi,

just to give you an idea of what could be wrong:
...the pump is only picking up the components some of the time...
Make sure the tube is not sharply bend or flow of air is otherwise disrupted on the way from pump to needle.

Make sure all connections are airtight so no extra air is "bleeding" into the vac system.

Make sure pump is supplied with sufficient current (weak power supply?) and produces proper vacuum.

If you have a vacuumeter (basically a negative manometer) you could check the suction pressure at the pump and at the luer lock for the needle.
-machine doesn't want to pick components up from the first reference hole on the tapes, it skips to somewhere on the middle for the 0805's even though on the tape positions table the X location specifies the right position. Though next X keeps increasing by about +30 every time I run a new job.
For each tape, LitePlacer "remembers" the last pickup position independently from current job. To start from first component again, you need to reinitalize (reset) the tape position or click "Reset all tapes".
-machine isn't placing the parts on the correct spot on the PCB. It varies, but seems it can be out by a good 3-4mm.
Make sure PCB coordinate (csv) file is correct and you have measured fiducials. You may click "go to measured location" in job configuration to see if camera moves into correct position for a specific part.

If this works, make sure to precisely calibrate needle - camera distance - so components are picked up and placed at the right spot. You may try "pick up this" and "place here" with single parts and manual movement to see if this works correctly.
-just ran another job and when I put the second needle on for the IC it went over the up camera and then cancelled, saying something like "can't see camera".
Check cameras are configured correctly and visual processing filters work for part detection.

Good luck!
tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Re: Vacuum pump issues

Post by tek »

I found that I had only 12v going to the vacuum pump, so I gave it 24v now it works.

I didn't realize the machine was saving the tape progress though I guess that makes sense. Unfortunately however i have another issue in that it is trying to pick the first component up to the left of the reference hole instead of right. That included trying to pick up non-existent components from empty holes. Also, sometimes it seems like the pump is not engaging on the first component, because quite often it won't pick up the very first component in the job (when there actually is a component).

I assumed the problem with bad placement was a badly calibrated needle. Could be, but I recalibrated it and as far as I can tell I got it spot on, but now it's even more inaccurate, so not sure what to do there.
mrandt
Posts: 407
Joined: Mon Apr 27, 2015 10:56 am
Location: Stuttgart, Germany

Re: Vacuum pump issues

Post by mrandt »

Good you solved the pump issue.

From the top of my head, I think from tapes the component left and up from sprocket hole is picked - so just set hole 1 accordingly in tape setup dialog.

For pickup errors, you should experiment with timing settings - basically wait time after engaging the solenoid. If problem only occurs for first component only, it might be that there is not "enough" vacuum created in the long tube from the pump to the solenoid just yet. Might help to enable pump through checkbox on "Basic Setup" first to evacuate the long tube.

For the bad placement, there could be a ton of things that go wrong. Without further details, impossible to diagnose remotely.

Make sure you follow Juha's guide for calibration literally step by step!

Other suggestions:

Most important is probably the calibration of camera to needle distance but if pickup works, that seems ok.

Check if camera moves to correct spot on PCB when you click go to "measured location" on Run Job (after importing CSV) for a few components.

Try manual pickup and placement (buttons on bottom right) and see if components end up in the intended spot.

If components require rotation, check if tapes are setup correctly, CSV file is fine and that stepper motor for A axis (rotation) is setup correctly - e.g. rotate by 180° using F-keys and make sure needle actually is rotated by desired angle.
tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Re: Vacuum pump issues

Post by tek »

From the top of my head, I think from tapes the component left and up from sprocket hole is picked - so just set hole 1 accordingly in tape setup dialog.
My mistake, I thought it was up and right, I've moved the reference hole across one, should fix that issue.
For pickup errors, you should experiment with timing settings - basically wait time after engaging the solenoid. If problem only occurs for first component only, it might be that there is not "enough" vacuum created in the long tube from the pump to the solenoid just yet. Might help to enable pump through checkbox on "Basic Setup" first to evacuate the long tube.
It WAS only missing the first one yesterday. It seems worse today, the last run I did it only picked up 2 resistors out of 6.

I filmed a job on my phone, it's not great quality but perhaps it will help?
https://www.youtube.com/watch?v=HVmDAD357Ls

As far as I'm aware I've followed everything step by step. I might try messing with the cameras and then going through all the calibration again.
JuKu
Site Admin
Posts: 1114
Joined: Thu Feb 14, 2013 3:06 pm
Location: Tampere, Finland
Contact:

Re: Vacuum pump issues

Post by JuKu »

Yes, something here is not as it is supposed to be, although I don't immediately see what. Does you drive the valve with 24V? Do you hear the valve operating when it should? Does changing the time has any effect? Does the "pickup this" function shine any light to the issue?
tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Re: Vacuum pump issues

Post by tek »

JuKu wrote:Yes, something here is not as it is supposed to be, although I don't immediately see what. Does you drive the valve with 24V? Do you hear the valve operating when it should? Does changing the time has any effect? Does the "pickup this" function shine any light to the issue?
-I am driving it with 24v, was 12v previously but that was an error

-I THINK I hear it as I should, though I have to admit I don't really know what I'm listening for:
The vacuum pump hums constantly when it's on, the needle lowers, the valve goes click, the hum sounds deeper, needle lifts. If there's a component, the hum remains deeper, if not, the hum goes back to normal. Carries (or doesn't carry) component to the board, lowers to the board, makes a pop sound, and then the hum returns to normal. Repeat. (I feel like an idiot for that description :lol: )
The IC is picked up 100% of the time but the resistors seem to be picked up about 40% of the time.

-I've changed the pickup time between about 250 and 800ms, doesn't seem to have any effect

-Using "pickup this" seems to have about the same failure rate.
JuKu
Site Admin
Posts: 1114
Joined: Thu Feb 14, 2013 3:06 pm
Location: Tampere, Finland
Contact:

Re: Vacuum pump issues

Post by JuKu »

> I feel like an idiot for that description :lol:

But it was spot on, and what I needed. Nothing wrong there, it is as it should. The pump and the valve are operating when expected.

> The IC is picked up 100% of the time ...

This narrows it down considerably. You have enough vacuum and the pickup system basics work, the issue is on small parts.

Is your needle calibration good enough and the needle small enough that the needle lands fully on top of a resistor, and not partially on the tape? Also, you could put a loose resistor on the table and test its pickup to make sure that the issue is not tape related. Although I don't really see how a tape could "hold on" to resistors, I guess somehow it could. I'll get back to here if I can think of something else.
tek
Posts: 15
Joined: Wed Apr 08, 2015 3:14 pm

Re: Vacuum pump issues

Post by tek »

Thankyou, I've been sidelined onto another job for the time being but I'll hopefully be back onto it before too long, maybe with someone else who might have some ideas.

I've noticed occasionally before I recalibrated everything that when the needle was slightly off, it would pick up the 0603's while only half covering them, causing them to be lifted at a strange angle. That hasn't happened recently, so I don't think that's the issue. The needle seems to be pretty accurate when picking them up, and you'd especially think that the 0805's would be less problematic, which I don't think they are. I'll consider trying loose pickups with some bigger resistors or other larger components.
Post Reply