2012 DOE grant upgrades

Pete Bachant edited this page Jun 26, 2016 · 10 revisions

This page is still missing some information, but will get finished eventually. -- Pete

In 2012, the tow system's linear guide, motion, control, and data acquisition systems were upgraded. Funding for equipment was provided by an infrastructure grant from the US DOE.

Problems addressed

  • Linear guides: The linear guide system, designed and implemented by Darnell in 1996 had failed structurally, and been patched with stainless plates held on by double-sided tape. These shifted around significantly during carriage motion.
  • Motion and control: Carriage acceleration was very low---on the order of 0.2 m/s². This severely limited usable tank length. Motion was controlled using an open-loop velocity control. Positioning was done by the user estimating coast time and removing tow power accordingly. Obviously this was not safe. Furthermore, large low frequency oscillations in tow speed were seen due to the highly compliant 0.25 inch diameter wire rope tow member.
  • Data acquisition and on-board accessories: Data was acquired by an on-board PC or laptop, which was controlled via Windows Remote Desktop over a Wi-Fi network. On-board power was limited to 4, 12-volt automotive batteries, which could not power the ME department's PIV system. Furthermore, a goal was to automate the turbine test bed, which would also draw more power than the batteries' capacity.

Concepts and implementation

Linear guides

The linear guide system was based around Thomson 1.25" diameter 440C stainless steel shafts. Existing quasi-level surfaces were used for mounting/adaptation to keep cost down, since a quote from Oasis Alignment for installing new level surfaces with the shafts went into the $85k range. We also saved money by doing the work ourselves (Wosnik's students Bachant and Lyon). The shafts are mounted on threaded studs in oversized holes to allow for adjustment in all directions--a concept borrowed from Saint Anthony Falls Lab (SAFL).


To accommodate the new linear shafts, Thomson pillow block linear bearings had to be installed at the four corners of the existing carriage. This was accomplished by cutting off the old hardware square with the carriage edges, installing custom designed adapter blocks on the master (longer, with 6" x 6" beam) side, and installing custom fabricated brackets on the slave side. The slave side brackets are attached to an 80/20 1530 extrusion, which is clamped to the carriage with 3 custom designed clamp assemblies. The pillow blocks mount to the slave side brackets in slots with the screws loose, to allow for deviations in parallelism between the master and slave shafts in the cross-tank direction.

Motion and control

The 0.25 inch diameter wire rope was replaced with a 7.5 cm wide steel-reinforced BrecoFlex ATL 20 timing belt. This belt runs through a custom-designed housing with symmetrical pulley housings at each end, driven at the wavemaker side. The drive pulley is mated to a Kollmorgen AKM82 servo motor via a 10:1 gearbox, which is powered by a Kollmorgen S700 servo drive. The drive is commanded by an ACS NTM standalone 8-axis EtherCAT master controller. This setup now allows for precise closed-loop control of the carriage's position and velocity, along with increasing maximum acceleration by over an order of magnitude to 2 m/s2, which is very important considering the relatively short tank length.

Meeting new requirements

TODO: Finish this section

Requirement Target Prediction method
Max tow speed 3 m/s Kollmorgen Motioneering software was used by Target Electronics/Minarik to ensure steady force based on 1 m2 submerged frontal area with a drag coefficient of 1.2 (5400 N) could be met by motor/gearhead combination.
Max acceleration 2 m/s2 Kollmorgen Motioneering software. See above. Additional inertial loading computed based on a 400 kg towed mass, bringing total max force during acceleration up to 6200 N.
Max tow force 6200 N Requirement derived from above. To verify chosen concepts would work, the following calculations were checked: linear shaft mount FEA (SolidWorks Simulation), belt tensile strength, belt tooth shear, belt tensioner bracket FEA (SolidWorks Simulation), pulley housing FEA (SolidWorks simulation).
Onboard power Run PIV system and onboard turbine servo. Single phase amperage for PIV computed based on component fuses. Onboard servo sized with Kollmorgen Motioneering software.


See the budget overview spreadsheet.


Things that could/should have been done differently

Drive member

I probably would have opted for a rack and roller pinion system to drive the carriage rather than a belt, for the stiffest drive system possible. This would have costed around $15k or so--a little more than the belt, but not too bad. The motor and drive would have to be attached to the carriage, which would take up quite a bit of space and add a bit of mass, but it might be worth it in the future, especially for towing objects that impart oscillatory forces, e.g., cross-flow turbines.

DAQ hardware

It would have been nice to have the motion controller and DAQ hardware synchronized over some real-time network, e.g. EtherCAT. It would probably be possible to replace the NI 9188 Ethernet chassis with an NI EtherCAT slave chassis, and have the ACS controller deal with data acquisition over the network, but this would probably require some sophisticated low-level programming to deal with the EtherCAT data.

Photo album

Note: This album is incomplete.

2012 UNH tow tank upgrades
<script async src="//s.imgur.com/min/embed.js" charset="utf-8"></script>
You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.