Spring loaded adapter affecting nozzle height calibration

[Timbosaurus]

With the new nozzle kit installed, the force required to compress the nozzle adapter spring seems to be about the same as the force required to push the tube up and trigger the z probe switch. Consequently, sometimes the nozzle spring compresses first, and sometimes the tube spring compresses first, so the z value at which the probe triggers is very inconsistant.

My findings are:
with no tube return springs connected (gravity return only), there is no nozzle spring compression before the z probe switch triggers
with the tube return spring giving the slightest amount of return force, the nozzle spring compresses halfway before the z probe switch triggers
with the tube return spring giving the most amount of return force possible, the nozzle spring compresses fully before the z probe switch triggers

So I can tune it one way or the other by adjusting the tube springs, but this will likely change as the bearing wear/change temps/get side load/etc.

I am not familiar enough with the operation to know how this will affect things, so can someone offer advice on whether I've done something wrong, or which way I should be adjusting the spring load? Or maybe the nozzle height results are not critical enough to cause issues?

Cheers,
Tim

edit:
just FYI, the needle height calibration results are:
no/low tube spring force 35.80/1.8
high tube spring force 35.52/2.9

and depending on how the spring and bearings feel like behaving, it may be anywhere in this range. So I can try and keep the preload at one end of the range as best I can, but will this variation cause problems?

[JuKu]

With the new nozzle kit installed, the force required to compress the nozzle adapter spring seems to be about the same as the force required to push the tube up and trigger the z probe switch.

Please try to tune the bearing/tube mounting. The force to push up the tube should be significantly less than to compress the spring in the nozzle holder.

[Timbosaurus]

Hi Juha,

Thanks, that does makes sense. The adapter force is only just enough to push the limit switch in though, let along the friction of the bearings etc. But I've retuned the bearings as best I can, and it is now fairly consistant on the nozzle calibration at least.

But I am still having a small issue with the Z-axis in the hello world program, and I'm not sure if it is related...

The 0805's work fine, and the first 0603 works fine, but when it comes to pick up the second 0603 (after auto z-pickup height is set), it says "the operation seems to take the nozzle below safe level, continue?"

The Z settings are :

Z0 to PCB = 35.32
Backoff = 2.2
Placement depth = 1.00

Tape Z's (after first component is loaded):
Pickup Z = 37.456
Place Z = 35.899

Command when it comes to pickup the second 0603:
ReadyEvent stat
{"qr":32,"qi":0,"qo":1}
PickUpPart_m(): Part pickup, Z37.456
==> {"gc":"M05"}
CNC_XY_m, x: 111.511, y: 212.506
{"r":{},"f":[1,0,13,72]}
==> {"gc":"G1 F150 X111.511 Y212.506"}

What is the safe limit? If I manually set pickup z to 37, it is fine, but it would be good to know why the auto sensed value is too low.

Or am I missing something more fundamental?

Thanks,
Tim

[JuKu]

There is a safeguard setting "Allow Nozzle to below PCB" on the basic setup page, above the Nozzle height box. The default value is bit small, 2mm. Your numbers say that it wants to pickup the part 37.456-35.32 = 2.136mm below the PCB level. Increasing the safeguard to 3mm or so should fix this.

[Timbosaurus]

Brilliant! Thanks!

[ocmaster]

With the new CP40 nozzle kit, just remove all springs on the pickup tube assembly, they can't be used, since just the slightest pre-preasure on the tube will make the CP40 nozzle holder move before the tube itself, resulting in the limit switch to trigger with false measurements, since the CP40 holder will compress at least a couple of millimeters before the tube starts to move towards triggering the limit switch..

We simply removed both springs and just left the collar on top of the top tube bearing, so it prevents dust from entering the bearing..
The trick to get the best movement is a couple of drops of silicone spray in the tube bearings, after this is applied, the tube will move completely free and won't need any springs to be pushed all the way back down when the nozzle is not in contact..

Off course the alignment of the two tube bearings are a a must before you will get a perfect non-friction movement on the tube..

I hope the above makes sense - we spend quite some time trying out different things - most of them without any luck..


@Juha - sorry for not replying your latest mail - things began to work, so we have been producing But the above solution fixed the issues with the Zmax limit switch.. !
It now runs like a dream..

[mawa]

As you can see from my picture in http://liteplacer.com/phpBB/viewtopic.php?f=3&t=382
I am using Juhas double spring solution which I seem to have intuitively adjusted right, as the z-Axis probe function worked immediately just the way it did with the old needle head.
Calibration of table height, tape and manual pickup and place operations work just like they did before the upgrade.

You do have to bring the z-limit switch very close to switching. When I push up the nozzle I get no adapter-spring compressing before the switch is activated and then pushing further up it contracts less than a mm while the tube also still is moving up.

Repeated calibrations of height returned very similar values.

So ocmaster I suppose you did not put enough force into the upper spring.

I would not rely on gravity to bring the nozzle + tube down to its lowest position because after first probing you pick and place the further parts of that tape with the probed height values.

[Timbosaurus]

so it seems that there is no perfect solution for adjusting the tube springs, as everyone has variation in bearing friction, CP40 nozzle force, and likely probe switch force/distance too.

So what I take from this is:
- adjust the bearings as best as you can
- use some lubricant on the bearings (i used sewing machine oil - which didn't do much, so I will try the silicone based stuff next)
- add enough tube spring force to ensure that it will overcome bearing friction and return down consistantly (i needed to use a bit of spring force - like manfred, but ocmaster needed none)
- adjust the z-probe switch for minimal travel before the switch triggers

The main aim is for the adapter spring to not compress at all before z probe switch triggers

Finally, make sure the lower collar is adjusted to hit the bearing after the z probe switch triggers, but before the tube spring fully compresses and before the z probe switch arm bends too much (for me, i need to use this collar to push against the bearing to provide enough force to load the nozzle during nozzle changes)

I have no experience with the luer lock needles, but this auto change system seems brilliant in comparison!

[mawa]

I have no experience with the luer lock needles, but this auto change system seems brilliant in comparison!

That is true! I can only thank Juha for the new nozzle solution.

The only problem I currently have is that two of the six nozzles have a significantly higher friction and need a lot of force to mount and dismount the nozzle adapter. : I have not setup the changer yet, but I fear the z-axis will need extreme force on these two nozzles.

Due to the fact that I ordered 4 adapters I can report that the 2 nozzles diameters are off standard size not the adapter.

[mrandt]

A bit off-topic:

Anyone knows how Mechatronic's height measurement works?

https://youtu.be/3kkf6bktIHA?t=37s

I am not sure if it is a mechanism in the nozzle or in the "metal" cube they use to determine component height...

Cheers
Malte