Tuesday, December 19, 2023

Redesigned Bus Bars

Original bus bar design


DSC01811 showing original bus bar configuration.  The bars are long and they pass over the full top of the controller.  After the main battery connections are made, then the B+ pre-charge terminal (top just left of center) is obstructed.   One defect to this design is the closeness of all of the terminal bars.



DSC02196 After all power connections have been made, this picture shows that the B+ pre-charge terminal is hidden under the bus bar that connects the fuse (white top center) to the B+
 controller terminal (center left).  


Second version with shorter bus bars


The redesign requires significantly less copper plate to make the bus bars and the top of the controller is much less congested, but the design requires that the polarity of the controller motor terminals be reassigned.  Controller terminal U1 must be changed to W1, and controller terminal W1 must be changed to U1.  The V1 center controller terminal remains unchanged.  A future post will step by step explain how the controller is reprogrammed.      


DSC02192   End on view of the new bus bars.  These prototypes were hand made and different vertical heights were tried.  The final design will be CNC cut and the U1 and W1 bars will be mirror images.  These bus bars use much less copper when compared to the originals.  The bus bars appear to be close together, but actually they are very separated as is shown in picture DSC02190 and DSC02189 below.  This is a dual shaft motor.


DSC02190 Close up top view showing how the U1 and W1 bus bars are offset (front to back) from the V1 bus bar.  The distance between the controller housing and the U1 and W1 bus bars is about half the distance between the controller housing and the V1 bus bar.


DSC02189 showing that the U1 and W1 bus bars are recessed and in a different plane when compared to the V1 bus bar.


DSC02195 New bus bars after placement of the controller outer box.  Note that the B+ pre-charge terminal (on bottom right of controller between the V1 and W1 controller terminals.) is no longer obstructed.  

Edited 12/20/2023

DSC02204 DSC02199 End view showing that both the motor terminals and the bus bars are fully enclosed within  the controller box by including a "drop down" leg in the design.  




Friday, September 1, 2023

Adapter Plate Revisit - Part 2

 

The 245 k bytes two dimensional DXF file that was created and previously described in the 6/30/2023, blog post was used within a CNC program to create the 2D outline of the housing.  That information was loaded and massaged in a program called Master Cam ( https://www.mastercam.com/  ) to yield a file that was then only 3706 k bytes.  The third dimension included at this point to define the depth of the internal and external cuts and the overall thickness of the plate.   Finally the Master Cam data was  loaded into a G and M Code that the CNC directly understands.   

https://en.wikipedia.org/wiki/G-code  

edited 9/5/2023

DSC01928 showing one of the previous adapter plates (from a casting) overlaid on the 15.875"x13.625"x2" billet of 7075 Aluminum.  


DSC02038 showing the trial "plate" produced from a piece of scrap 0.067" thick 6061 Aluminum using the DXF file after it was modified as described above to be compatible with the CNC machine.  The transmission is shown  standing on its drive shaft and the view is into the transmission housing.  This trial plate confirmed that the outline and hole locations were all correct prior to cutting the expensive 2" thick billet.


DSC02039 showing a detail of how precisely the DFX file conforms to the actual outline of the transmission housing.


DSC02041 of the Leadwell V-40 CNC machine that was used to manufacture the billet plate.

These machines are available on eBay and other resale platforms for $27.5 - 40K.  Some tooling is included.   https://www.leadwellcnc.com/  

IMG_9613 showing the billet loaded into the Leadwell CNC prior to machining.  Edited 9/5/2023


DSC02043 showing the initial cuts being made on what will be the transmission side of the future adapter plate.


DSC02073 Hyper9 (drive) side of the final billet adapter plate. The hole at the 12:00 is the center top and the hole at 6:00 is the center bottom of the transmission.


DSC02075 oblique view of the Hyper9 (drive) side of the final billet adapter plate.


DSC02072 Getrag transmission side of the final billet adapter plate.


DSC02076 oblique view of  the Getrag transmission side of the final billet adapter plate.


DSC02080 compares the transmission side of the original adapter that was made as a casting (left) to the billet adapter (right).  The casting weighs 4.92 kg and the billet weighs 5.66 kg.  The original casting is several years old and it has air oxidized and discolored in the shop environment.

 

DSC02078 compares the Hyper9 (drive) side of the casting adapter (left) to the billet adapter (right).  The casting adapter has more mounting holes to also accommodate either the Warp 9 or Warp 11 DC motors.  The casting hole pattern did not position the top center position of the transmission and the top center position of the Hyper9, so the 8 hole pattern closest to the 4" center opening was rotated slightly to the left.  The new billet hole pattern allows for the attachment of the Hyper9 mounting platform (DSC01563 of 2/23/2023 post) and the Hyper9 controller enclosure (DSC01830 of 5/20/2023 post) such that each are now level and parallel to the floor.  The holes in the billet at 1:00 and 7:00 are the top center and bottom center of the transmission respectively.

updated 9/2/2023

Sunday, August 6, 2023

Determining the Vertical Orientation of Getrag Transmission Relative to the Adapter Plate

 When making the previous adapter plate for use with a Warp11 series NetGain motor, the rotational orientation of the adapter relative to the motor was not critical as the power cables exited on the side of the motor.  By contrast, with the Hyper9, it is desired to locate the three terminals at the top of the motor, and all in a plane parallel to the floor.  If so located, then it will be possible to attach a controller or controller housing box directly on top of the Hyper9.  



DSC01947 Picture of a the Getrag 5 speed transmission installed in the E36.  Note that the seam in the transmission casting seam is exactly at the bottom and center when installed.  The center part of the cross piece is parallel to the floor, and thus perpendicular to the transmission seam.  Note that the mounting points on the right are higher then on the left.

To mimic the installation of the transmission, a spare gear box suspension cross piece (BMW OEM  23701136498, $58.93 on eBay) along with a pair of rubber bearings (BMW OEM 22316799331, $21.52 on eBay), were then attached to the transmission tail.  Two pairs of fully threaded bolts allow the transmission to sit squarely on the floor.



DSC01970 Top view of the rear transmission  suspension cross piece.  Top of picture is toward the front of the car. 


DSC01971 The fully threaded M12-1.75 ( L=  40 mm left and 90 mm right) bolts allow the transmission suspension cross piece to be leveled when resting on the floor.


DSC01978 Showing the adapter plate attached to the transmission on the left and the tail of the transmission mounted on the suspension cross piece on the right.  


DSC01980 Close up of the mounting.  The 4 bolts of the suspension cross piece permit leveling of the transmission from each direction.  


DSC01988 Showing the location on the current adapter plate where the top of the Hyper9 will be located.  When the Hyper9 is attached the center terminal will be in this plane. 

The Hyper9 has four mounting holes that relative to center and vertical are located at 45, 135, 225, and 315 degrees.

DSC01967 of the serial number for the transmission for the transmission being used.

This transmission 4USCC7328SLA01518  (2 121 643) has a slightly smaller outer contour when compared to the existing adapter plate.  This was also the case with the other Getrag transmission that was used. 



DSC01968 The bolt pattern is identical but the transmission contour is smaller then the adapter plate..  



DSC01969 Showing the slightly smaller outer contour for the entire periphery.  Bolts used for the pair of dowel pins is M12-1.75 x 90mm.

Friday, June 30, 2023

Adapter Plate Revisit 2023 - Part 1

 

Hexagon Absolute Arm 8525 7-Axis 2.5 Meter with AS1 Scanner

After talking with Dana Green at Exact Metrology (office 262-533-0797), traveled up to Brookfield, Wisconsin with the Getrag 5 speed transmission and watched as it was scanned.  The laboratory time was a little more then an hour, and the DXF file was returned to me the following day after the computer work had been completed.  Total charges for a simple 2D scan was $500.  

Exact Metrology

20515 Industry Ave

Brookfield, WI 53045


New location as of 9/1/2023 is:


3811 N Holton St

Milwaukee, WI 53212




DSC01918  Picture of  the Hexagon Absolute  Arm 8525 7-axis 2.5 meter with AS1 scanner.  The arm was secured to a large 7800 pound steel table.  These tools with software have a list price new of  $90-95,000 for a turnkey system.  This picture shows the beginning of the scan of the Getrag 5 speed transmission housing face.  The number (946273) in the top left corner of  the monitor is the number of data points collected so far.  




DSC01922 of the hand held AS1 Scanner.




DSC01919  The blue line is the actual data point collection area, and the red line outlines the window within which the blue line must reside for data acquisition.  The scan is a series of smaller "paint strokes" that the computer merges to create the 2D or 3D data set.  The final scan accuracy is expected to be +/- 1.5/1000" ( which is +/- 0.0381 mm).



DSC01921 Showing the green area of the fresh data points just collected, the grey-black area is of those points already logged, and the blue area shows the back side surface points for the collected data. 

The DXF file will next be given to the CNC shop and the adapter thickness and Hyper9 motor orientation added to create a 3D file.  Two adapter plates will be machined from a pair of 7075 Aluminum billets that are 2" x 13.5" x 18.75".  (ALRO Outlet in Elk Grove Village, Illinois at $185 each, or $3.49/ pound).

Note:  To see the previous posts regarding the efforts to make previous versions of the Adapter Plates, enter "Adapter Plate" into the "Search This Blog" window at the right of the page that includes the most recent blog entry.  

Monday, May 29, 2023

Internal Shelf Within the Hyper9 Control Cabinet

 A piece of 5052 Aluminum 0.061" x 26" x 27" was purchased at ALRO Metals Outlet ($14.00),  A 10.5" x 6.25" section was cut (DSC01847), and then formed into a platform with a top area of 10.5" x 3.25" and a height of about 1.625" (DSC01853 below).  The Gigavac contactor and the Shawmut fuse will be secured to the platform, and both will be significantly above the chiller plumbing.

DSC01847 showing the dimensions of the 5052 Aluminum plate prior to cutting with a bandsaw.

DSC01853 showing the end view of the formed platform.  The height to the top of the platform is 1.625".

DSC01855 showing the pair of slots that were cut in the platform to allow for the chiller coolant pipes.  The platform is upside down in this picture.

DSC01859 after placement of the Aluminum platform.

This platform fit easily between the cabinet wall and the controller (DSC01859).  The height of platform was the maximum possible that would allow a simple Copper 90 degree angle to be fabricated and used to connect the contactor and the bottom terminal of the main shutoff switch. 

Copper Bus Bars Internal Connections

A bandsaw, 12 ton press, and an Eastwood brake, were used extensively to produce the bus bars required to interconnect the "Power Input to the Master Shut off to the Gigavac Contactor to the Ferraz-Shawmut 500 amp Main Fuse to the Controller B+ Terminal".  All bus bars were of 0.125" x 1.0" wide Copper plate.  The contactor and main fuse were mounted to an Aluminum platform that will be described in the next post.  The height of the platform placed the contactor, fuse, and controller terminals all in the same plane.

DSC01869 of the power input B+ (top right) bar connecting to the top half of the main shutoff switch (top left).  This bus bar was previously described in the May 18, 2023 post.  


DSC01867 Bus bar to connect the master shut off switch B+ to the contactor prior to placement.  The piece was made from a piece of 0.125" x 3" long x 2" wide Copper plate that was bent into a 90 degree angle with the press brake.    

DSC01868  Bus bar connecting the master shut off switch B+ to the contactor (middle back of picture) after installation.  Removal of the Copper bar adjacent to the nuts (lower right) allowed for the use of a 17 mm socket wrench for tightening. 

DSC01860  Bus bar for connecting the contactor to the main fuse prior to placement.

DSC01861  Bus bar connecting the contactor to the main fuse after installation.

DSC01862  Side view of bus bar connecting the contactor to the main fuse after installation.  By taking advantage of the same height for the surfaces, no bends were required in the bar.

DSC01873  Side view of the bus bar Ver. 1 connecting the main fuse output to the controller B+ terminal prior to placement.

DSC01875 Oblique view of the bus bar Ver. 1 connecting the fuse output to the controller B+ terminal prior to placement.

DSC01871  Bus bar Ver. 1 connecting the fuse output to the controller B+ terminal after installation.  The overall height above the mounting points is 1.25".  Note that there is easy access to the B+ pre-charge terminal underneath this bar.     

edited 5/31/2023

DSC01888 Bus bar Ver. 2 connecting the fuse output to the controller B+ terminal after installation.  Making the vertical height and the tab length at 1.25: fit the press brake well, but a 1.0" height and 1" tab length was difficult.  The use of a vise and mallet was required to finish the job.


DSC01889 Bus bar Ver. 2 connecting the fuse output to the controller B+ terminal after installation.  No loss of access to the B+ pre-charge in this smaller lighter version.  The overall height is now 1.0".



DSC01876 Showing the close proximity of the controller's K3  RS-232 serial port to the Power B- terminal on the cabinet (bottom right side).   

Edited 5/30/2023

DSC01878 Showing the manila folder cardboard test model that was used during the design process  (the wooden rod for additional rigidity).


DSC01884 Showing the preliminary Aluminum test model fashioned after the cardboard test piece.  Note that there is significant clearance around the RS-232 serial port 


DSC01883 Detail of the connection at the power B- input terminal on the side of the cabinet.


DSC01887 Showing the vertical height of the Aluminum test bus bar above the U1,V1, and W1 controller terminals.  The final bus bar will next be fabricated in Copper.