Strange registration issue

[ZooCity]

I am good to above average in my troubleshooting skills with presses but we had an issue on a print run yesterday that has blown my mind.  Here it goes:

Sportsman EX 2013, appx 165k impressions, installed and registered by M&R tech last Dec, air locks, sq air pressure.
6 color plastisol job: UB - Red Chili flash @ 100power and 2-3s - Top Color - Top Color - Top Color - Reno flash @ 2-3s - Hilite  Pressure on all heads low as possible to clear.  All 225/40 s mesh, all single strokes.
Smaller, high back "locker tag" location, ≈2" down from bottom of back colllar seam. 
Sq drop/front prox set to hit at platen edge.  Collars loaded off the platen.  ≈2-3" of lead out at end of stroke.  Standard M&R 16x22 platens with rubber top.
Warm, tacky platens
Level press with mfg maintenance schedule up to date.
Micros zeroed out after adjustment, no torsion on the frame as I regged this job and my m.o. is to unlock the rear frame holder. 
≈2k pc run, operating at 63 doz/hr

The UB was jumping out of reg like nothing I have ever seen. We regged this job over 5 times over 8 hours and each time we began running, about 15 shirts in it would jump out over 1/16" of an inch.  We burned a new UB with even more choke, though the first UB had adequate choke on it.

We checked everything, I mean every swinging thing you can think of on the machine.  In the end, I unlocked the rear clamp on the UB, moved it so that the clamp was snug up against the back of the screen frame and locked it in.  Then I snugged down the manual clamps as well.  Regged one last time and we were finally able to run.  However, each print was different, no pattern at all to where UB the printed.  The choke allowed us to run the job but that UB was trying to peak out the entire time.  We had no other issues and the top colors all stayed in registration or a least acted "normally" needing only minor adjustments if any across the print run.

I did contact M&R to help troubleshoot in case I'm missing something.

We are slammed right now as usual this time of year but the next step is burning a set of registration screens and testing to try and narrow down what is happening. 

I have seen this style of press bounce out of reg on a head but have almost always been able to track down why.  With identical screens and similar pressure and angle on all heads I'm hesitant to blame it on deflection and as I mentioned we're already printing as light as possible.  The problem repeated on a different screen with the increased choke.

I fear that it is a registration issue but I can't see how that's possible a year out from install on a near new machine with such low impressions on it between then and now.  Anyone have an idea or another angle of attack? 

[ebscreen]

Out of reg left-to-right or top-to-bottom or all of the above? Machine reg won't be top to bottom unless your
center bearing is going.

Can the heads themselves come loose?

What fabric?

[cleveprint]

not sure how the Sportsman is built, but one time on our diamondback, one of the pallet arms came loose. the bolts holding it to the carousel came loose and one screen kept coming out of registration. it took me forever to figure it out. cranked them down tight and it was business as usual. but again, I'm not sure how the sportsman is built and it could be different.

[ZooCity]

All over, it was just walking around.  Great point about machine reg though.

M&R tech pointed out to check for wiggle with the head flipped up which I'm definitely going to look at, it's the one place I didn't think to check.   So I guess maybe that part of the head could loosen.

Ringspun cotton. 

We're running OT here since last week but I'll crawl in there and check all the bolts more thoroughly.

[jvanick]

how often do you grease the center bearings?

-J

[ZooCity]

how often do you grease the center bearings?

-J

On the schedule in the manual.  We have a very thorough maintenance calendar/task setup for all equipment and the production crew keeps up on it as far as I can tell but i'll double check to be sure.

[Underbase37]

Rear micro's not locked down? I'm sure you would have noticed this but just throwing it out there.

Grey shirts?

Are you using the press today without troubles.

Murphy37

[GKitson]

Try this scenario, if the garment has a sewn in tag and/or your glue/tack is irregular at the edge of the platen you will see this irregular jumping around of registration caused by the floodbar and or squeegee pulling the garment during the print stroke pass.  It will drive you friggin' bonkers.

Our printers mechanical mind goes right to press registration when I have seen this with irregular glue deposit.

It may also be that a printed or heat applied tag does not bed in the glue tack the same way as the fabric and 'moves' during printing.

My 2 cents....

~Kitson

[ZooCity]

Lord I wish it was one of those things but we tackled all that.  Light green/mint shirts. 

Tack and having the collar or shoulder seams on the platen was immediately addressed and I agree we printers tend to overlook the basic adhesion of the shirt to the platen sometimes.  When I have issues like this with no pattern to it I go first to the substrate.  The shirts are easily the least consistent variable in the whole equation.  We printed probably over a hundred testers with all sorts of different fabric, seams, tags, used testers of the identical shirt style as well and still there was no rhyme or reason to the misprinting.

The print stroke speeds on this job were very slow for us, 3 or 4 on the dial, due to the image being right up top of the shirt and needing to give the squeegee blade real estate to start in and "recover".   This has me looking at deflection again, as the heads deflect up (platen arms are pretty strong on this machine, heads do flex up significantly) the extended time in the deflected position may be part of the equation.  But we've ran many similar jobs with no problems so I'm no sure about this theory.

Lots of DC on the press today and nothing printing in head 1.  Next job up is HSA with a UB in head one so we'll get to check it then.  I honestly think it would mess with me more if the problem up and disappeared today...

[jvanick]

how much pressure are you running?  -- we only see the head deflection over 50psi.

[DannyGruninger]

how much pressure are you running?  -- we only see the head deflection over 50psi.

I can't answer for chris but I will say I know he uses a lot of thin thread/barely enough pressure to clear the mesh.... I am 100% confident knowing the type of printer/owner he is that he's not overlooking something as simple as pressure, glue, etc...... Chris is the type of guy that is leveling pallets every week, maintaining his stuff per manf. specs, etc....... Just wanted to throw that out there LOL..... I'm trying to wrap my head around this issue as well and right now I'm blank, sorry! LOL

[UnderPressureSP]

I would check every bolt on head one to see if anything is lose and tighten .   Also the next time you print with that head run a long round of test prints 50+ and take a long look at that head while the machine prints.  My guess its was a setup error and something like the mircos were not tighten all the way or some bolts were loose some where.  In the future to avoid deflection issues place the shirt collar lower on the platen and use a laser or tape for collar placement.  I always hate when I start seeing reg issues of the press.  Also check your screen clamps to make sure their set right and no loose bolts.  And if a bolt did come loose.....  was it put back together right.  Good luck but I bet it was user fault something that will melt away and resurface another day......  Unless its a bolt coming undone and undone again after tightening.   I bet it was the glue 1st screen is heavy and pulls up a little bit and 2nd one has less ink and is flashed and pushes down the shirt into the tack.  Watch learn and repeat. 

[ZooCity]

Ha, thanks Danny, I feel bad for bringing this up and letting it get on anyone else's mind, it's like a brain infection that will drive you mad. 

I haven't been in a lot of other shops but from those I've seen we definitely put more time and energy into PM and doing things the way the mfg intended, it's the only way to have that baseline to troubleshoot problems from in my opinion.  Also, there are times when machines do fail and I like to know for sure that's what the actual problem is. 

That said, we're not perfect by any stretch of the imagination and I wouldn't put it past our shop to be missing something obvious.  It's the real reason I reached out on here, that chance that someone would point the most basic thing out that we were derping out on the whole time.  I'm still hoping it winds up being something simple.

So I got excited to get a machine in the shop with adjustable and measurable psi for pressure on the print stroke last year.  I really thought it would help in comparing notes with others and in our own setups but what I've found is that tossing out this or that psi gets irrelevant pretty quickly.  If discussing different machines, don't bother comparing psi.  Same machines, it's still tough, so many variables change the scenario, as we all are familiar with.   I think there was a good thread where this came up earlier this year.   That said, I'm always on the quest for printing with even less pressure, there are many benefits to doing so, machine wear and tear being a major one in my mind.  We do see significant upward deflection of the print heads at psi well below 50 but again, I'm not sure how relevant that figure is.

And I should remind that I don't print very often in our shop anymore so I'm only out there when everyone up through my PM has hit a wall with a job, I see all of the worst and miss most of the best runs. 

[californiadreamin]

With Absence Of A Registration Device I would Print An X-Y grid Sheet On a minimum of 3 colors to check registration.
Check the print pallet to pallet and see if it drifts.
If it is on registration on a number of prints and then it is not, it is your machine registration.
IMHO someone of your caliber, should also be registering your own press. You would pick it up Quickly! It is Easy.
winston

[ZooCity]

With Absence Of A Registration Device I would Print An X-Y grid Sheet On a minimum of 3 colors to check registration.
Check the print pallet to pallet and see if it drifts.
If it is on registration on a number of prints and then it is not, it is your machine registration.
IMHO someone of your caliber, should also be registering your own press. You would pick it up Quickly! It is Easy.
winston

Great plan!  Given our location, I should probably buy the device to do this for our presses.  Aside from warranty installs, press registration is the one item I need a tech for and getting a tech out to my location is not cheap.  I get the math behind the process and the procedure, at least I think I do!  Thanks for the push.  What tool do you recommend I, or anyone with the right skill set to do this, pickup for M&R style machines?   

All this with the disclaimer that registering a press is not for everyone and always best left to a tech if you have one readily available of course.