Tension Problem on Paper Converting Machine

I don't know your exact setup, but when I was in this industry, the tension load cell inputs were to amplifier cards. Sometimes the cards needed to be rezeroed and respanned. Sounds like the "zero" is off. Have you checked these (or your machine's equivalent)?

The tension at the roll sensor is just dropping some of the tension, but is staying in an acceptable range. The main loss in in the mechanical end, between embossing rolls and pull rolls.

What's the mass on these rolls, your dancer arm may need to be adjusted in the program to account this condition when the machine is stopped.

or you may need to add a hysteresis brake to keep the tension.

Happens with full or almost empty roll. It is from a new paper manufacture we are getting our paper from now.

I have absolutely no knowledge about the operation of paper machines, but my oldest brother who has spent more than a few decades in the business looked at this posting and emailed me back. He suggested that maybe the drive motor is not in the stall tension mode.

I would also suggest that there may me a problem with a sensor that is activated at <stop> that tells the PLC to retain the last tension signal from the load cell to the PLC.

I can't be more specific since I don't know anything specific about your setup! <hint, hint>

Are the gears between the embossing and pull out rolls worn or a cog slipping on a shaft for part of a revolution (worn key/keyway)? When the machine is running, are all the tolerances taken up in one direction? Which would happen if the pull out is powering the emboss via the gearbox? When the machine stops is there sufficient inertia on the emboss for it to over run and take up all the tolerances in the other direction and create the slack? To check this simply, put a wedge between the pull out and the machine frame to lock it into position, then try to rotate the emboss forwards or backwards.

Disclaimer: I know absolutely nothing about this sort of thing.

Does the drive for the pull rolls remain engaged, or, does it dis-engage?

Is the new paper "more elastic", or, "less elastic" than the old paper?

Could you fit a simple tensioner between the embossing rolls and the pull rolls?

I'm struggling a bit to see why this matters. Do you waste a lot of product when you restart the process?

Hi, As you are running with the nip open at the callender rollers the tension will only be controlled by the dancer, as the paper can slip over the polished roller. I assume the embossing roller and pull nip are geared together mechanically, or are they two separate motors, one acting as master and the other a slave? Are you trying to stop the machine too quickly? This new paper may not be as "elastic" as the previous paper you were using, so this fault may have always been there and not been noticed. Paul

When the machine comes to a stop does it run back slightly. This would cause low tension or even a slack web, due to the pull off nip running slightly faster than the embossing roller

is there an electric brake?, just replaced one on one of our wrappers.

a Schematic would help, otherwise we are just throwing stuff out there and then see what sticks.

Not anywhere near my specialty, but it seems as if the final stages are stopping fractionally faster than the leading stages, so the raw material feeds into the later stages giving you the "no tension" outcome

Since you are experiencing a tension problem only when you stop the machine, how about not stopping the machine? Problem solved

I understand that this problem has appeared with a different raw material, and no real other change, except that maybe something is broken.

I suppose it is possible that your new raw material is less elastic than your standard material, has to do with mostly with bulk, and some to do with the web tensile strength, and fiber orientation. The more the fibers are oriented in the cross direction, the more the web will stretch in the machine direction, without breaking.

There should be a Draw adjustment (hand wheel?) on your mechanical drive, to set the tension between the embosser and the pull rolls. It is possible on the previous web, there was enough elasticity in the web to hold tension between running and stop conditions, and maybe the draw adjustment could be tweaked?

We used to have plenty of converting machines. It would be good to inform us what drive type is on yours.

The dancers are supposed to control the brakes and speed. Are your brakes dirty or worn out? Perhaps with adjusting the springs tensions you might find a way out.

We had machines up to 3 m wide and running at 240 m/min. Drives were hydraulic, some servo electric, and some fully PLC controlled with DC motors.

To help you out we need more info about your converting- embossing machine.

Mass, diameter of rolls, is vacuum suction applied on the rollers? The accumulator capacity to store the loose paper etc...

Perhaps some pics?

Are the dancers 100% speed control or only a trim. If the drives are not matched properly the trim dancers may not be able to correct for speed variances.

Your issue may be with control code. If each drive is PLC controlled, then in the code you need to decelerate the first station and program the rest to continue to drive based on the tension sensors and not fully stop until you have a clear zero speed indication at each station and stable tension between.

I am unclear what actual quality problem you are trying to solve. is there a registration issue between stations that causes product rejection, or web fails or wrinkles that cause product rejection?

Where the issue showed up with a change in input material, do you have any objective quality tests to determine paper properties that affect the tension, such as compression, elastic and plastic deformation of the paper at process strains and loads or creep of the paper in tension over time?

Ideally you would like to set up the process so that input properties don't affect the output, but in most cases I see that the input material properties aren't identified and the effect on processing isn't known.

Jim, CSSBB