From: "Larson, Rod" <Rod.Larson@mts.com>
To: "Brad Thoen" <brad@thoen.com>
Subject: RE: Questions regarding old static support leveling (MORE)
Date: Monday, March 31, 2008 5:11 PM

I think your idea of time slicing is NOT a bad one. I will try to remember auto sequencing that I did at Saclay. It was 3 years ago...hehehehe. BUT, I am pretty sure that it would NOT take 40 minutes under any circumstances.
 
We are trying Lateral control today at NCREE.......
 
Rod
 

   _____  

From: Brad Thoen [mailto:brad@thoen.com] 
Sent: Monday, March 31, 2008 2:35 PM
To: Larson, Rod
Subject: Re: Questions regarding old static support leveling (MORE)


Rod,
 
Forgot to reply to your remarks in 2) below:
 
The current automatic static support leveling method already has a vertical force setpoint:
 

 
The current automatic leveler does in fact use DOF force control:  It does not try to zero actuator delta-P.  Instead it tries to zero DOF vertical force (minus the setpoint) and roll and pitch moments.
 
-Brad
 

Brad Thoen
MTS Systems Corporation
Advanced Systems Division
14000 Technology Drive
Eden Prairie, MN 55344 USA
O: +1.952.937.4244
C: +1.952.448.2004
HYPERLINK "mailto:brad@thoen.com"brad@thoen.com

----- Original Message ----- 
From: HYPERLINK "mailto:Rod.Larson@mts.com"Larson, Rod 
To: HYPERLINK "mailto:brad@thoen.com"Brad Thoen 
Sent: Saturday, March 29, 2008 5:22 PM
Subject: RE: Questions regarding old static support leveling

Good questions...
 
1. Depending on the system, the air pumps usually were quite small; therefore, it took a long time to get the pressure up. The exception was CABR. That customer put in a really big air compressor and it could charge up DAMN quick (few seconds). I would say that an accurate leveling at Azalee would take about 10-15 minutes. Most of the time was in re-selection of bags when you got close to done. In the early part, it is easy. Turn them all on for a few minutes. But, in the end, you would run an average of 10-30 seconds and then re-select. The manual approach was a REAL pain-in-the-ass for the operator. If the load is not centered on the table, the manual approach took even longer. I would say that Azalee was in the 30-45 minutes time frame then.
 
2. The automatic systems work FAR better. I think that it takes somewhere around 1/2 to 1/3 the time. BUT, We NEVER want exactly zero vertical load. Why? Because in an e-stop situation, the table might NOT go down gracefully or worse yet, the table will park with remaining hydraulic pressure only to pop back up weirdly after oil pressure gone. So, it is important to have a setpoint that allows slightly positive forces (table will stay down).
 
I would recommend that we continue to use automatic leveling w/ an average setpoint for the total static support. Why just 1 setpoint? cuz then you can handle an off-center load quicker. The computer should do the off-center load adjustment. Get each vertical actuator load to be equal no matter what is on the table (requires different pressure in each bag). Use a DOF kind of control approach. In actuator load.....Roll force = 0, Pitch force = 0, and vertical force = setpoint. The resulting bag pressures, of course, will be unequal for off-center load, but the table forces from the actuators are balanced.
 
There is another argument. Because of the greatly improved force balance that you use now, Perhaps it is better NOT to make the actuator forces equal. Make the force balance handle the off-center load. This is what I did at CABR and I think it was better to do that instead of making all actuator forces equal. CABR static support was run by the PLC and could have individual setpoints, so I could try both methods. The new force balance would produce the torques necessary to keep the table flat and that seems to be a better operating point from which to start dynamic operation than the equal force condition.  Something to think about?????
 
Rod
 

   _____  

From: Brad Thoen [mailto:brad@thoen.com] 
Sent: Saturday, March 29, 2008 9:56 AM
To: Larson, Rod
Subject: Questions regarding old static support leveling


Rod,
 
I have two questions about the old static support we used to provide.  CEA Azalee and the old SUNY table have this style.  The static supports are airbags, and you select which bag(s) you wish to charge or discharge, the you charge/discharge all of the selected bag(s).  This is different from the new style static support in that the new one allows individual charge/discharge, i.e., you could be charging one static support while discharging the other at the same time.  Also, the old style are levelled manually, whereas the new style allows auto-levelling.
 
My questions regarding the old static supports are:
 
1) A question regarding physics:  From start to finish, how long does it take to level the static supports, assuming you manually did it perfectly the first time with no iteration.  In other words, how fast can the air pump charge the static supports?
 
2) A question regarding skill:  How much longer than the time in (1) above does it take a reasonably skilled operator to iterate to a good levelling solution?
 
Thanks.
 
-Brad
 
Brad Thoen
MTS Systems Corporation
Advanced Systems Division
14000 Technology Drive
Eden Prairie, MN 55344 USA
O: +1.952.937.4244
C: +1.952.448.2004
HYPERLINK "mailto:brad@thoen.com"brad@thoen.com


No virus found in this incoming message.
Checked by AVG.
Version: 7.5.519 / Virus Database: 269.22.1/1349 - Release Date: 3/29/2008 5:02 PM



No virus found in this outgoing message.
Checked by AVG.
Version: 7.5.519 / Virus Database: 269.22.1/1349 - Release Date: 3/29/2008 5:02 PM




   _____  




No virus found in this incoming message.
Checked by AVG. 
Version: 7.5.519 / Virus Database: 269.22.1/1346 - Release Date: 3/27/2008 10:03 AM



No virus found in this incoming message.
Checked by AVG.
Version: 7.5.519 / Virus Database: 269.22.1/1352 - Release Date: 3/31/2008 10:13 AM



No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.519 / Virus Database: 269.22.1/1352 - Release Date: 3/31/2008 10:13 AM
 
  
 