Slip ring design for RMD-X7 motor board

[QuanLe8191]

No description provided.

[QuanLe8191]

https://www.directindustry.com/prod/jinpat-electronics-co-ltd/product-144799-1583175.html
Reference design for slip ring

[QuanLe8191]

https://www.aliexpress.com/item/4000381777381.html

[QuanLe8191]

@Jtayama54 Make some digging about these questions:
What is the dimension of the slip ring? How big it is? How many current do we need to deliver?

[kammce]

I don't understand the title of this issue. It is very confusing.

[QuanLe8191]

[HaadiElahi]

Slip ring dimensions based on trace widths/spacings for CANL, CANH, VCC, & GND.

in mm

[kammce]

A Few Additional Requests

Add HOME Pad

A discovery about the encoding scheme for the RMD-X7 motors shows that the encoder is not reading from the output of the motor shaft, but from the internal input motor. Meaning that 360 degrees of rotation of the encoder accounts for only 60 deg rotation on the output. But after 360 degrees of rotation, the count resets. During normal operation, the motor's software can track when it crosses between 360 degrees and back to 0 degrees, but when the motors are turned off, it loses track of that data.

In order for the drive controller to know where the motor is, we need to give the motor some sort of feedback signal indicating that it is at a predetermined position, a HOME position.

I request that a HOME pad is added to the slip ring on the outer race position. Just a small square or circular pad. On the opposite side, a pogo pin for the 5th race should be connected to GND. This way, the drive board can pull up the HOME pin and when it sees it fall to GND, it knows it hit the home position.

See this presentation for more details:

• https://docs.google.com/presentation/d/1f3TRtCSIvBHE58bGw5FQUc8k5HU7_gxTPD1gcxIJRI8/

Mounting holes

This design doesn't yet have a concept of mounting holes. I'd imagine that there would be 3 or 4 mounting holes on the PCB that allows for someone to screw one of the boards into the servo chair and the other into the drive fork. This is now even more important as the HOME pad needs to be aligned to the same position for each assembly and disassembly. If there aren't any mounting holes to keep the pad properly aligned

[QuanLe8191]

I have managed to create a circular trace in Easyeda. However, I think that we can definitely increase the trace width of the signal to increase connection with POGO pins as long as they all are spacing correctly

[QuanLe8191]

Maximum height: 3mm
Outer diameter: 100mm
Inner diameter: 91mm

[HaadiElahi]

Updated dimensions

Inner Diameter = 93 mm
traces = 53 mil
spaces = 15 mil
Outer Diameter = 111.034 mm (redundant dimension)

Note that the inner-most trace will be divided into 8 equal segments.

[QuanLe8191]

the innermost trace should be divided into 6 segments. Where is the mounting hole going to be?

[kammce]

@HaadiElahi I don't think you need to worry about designing the actual PCB in CAD. Also the traces are way way too small for anything the team will be doing. It may be best to simply have a simple placeholder in your CAD and let @QuanLe8191 do the CADing of the cirucular traces + HOME pad.

[QuanLe8191]

I make some change to the trace width:

• GND: 2mm
• VCC: 2mm
• CANH: 1mm
• CANL: 1mm

https://easyeda.com/editor#id=5334bd052eb2479e9ef1160dbc7233c1

[QuanLe8191]

Here is the 3D view of the top layer:

[kammce]

Why are they so thin? Its better that we make them larger so that we don't have to be super precise about the pogo pin locations. and about vibrations kicking the pogo pins out of place.

[kammce]

@QuanLe8191 Your easyeda link is giving me permission denied when I try to go to it.

[kammce]

Gold Plating The Exposed Traces

It is very important that we use gold plating for the exposed traces (or something close enough but cheaper). Tin plating is the standard and it is pretty useful more almost all uses to prevent oxidation and damage to exposed traces. But for the slip ring that will have have to endure continuous dragging of the pogo pin, using a gold plated (or strong plating) for the traces is something we should definitely do. Also the pogo pins should also be gold plated as well to make sure that they do not rub away as well.

For some context RAM slots and slotted devices in general use gold plating to prevent them from chipping.

[QuanLe8191]

Im thinking it will be quite difficult to do the tracing for the 6 fragment circle home
Add ADC schematic for the 6 fragment trace

[kammce]

Im thinking it will be quite difficult to do the tracing for the 6 fragment circle home
Add ADC schematic for the 6 fragment trace

I'm not sure if I understand what you mean here. Does this mean that you are going to revert to the single PAD home rather than the 6 sector analog pads.

[QuanLe8191]

Gerber view and price:

[QuanLe8191]

Fix the slipring dimension: currently the radius is 110mm (should be half)

[kammce]

Key Elements of the Slip Ring Design

• Contains 4 Conductive Circular Tracks:
  • VCC
  • GND
  • CANH
  • CANL
• Contains 1 Conductive Pad For Homing:
  • Shape: Square
  • When pogo pins on lower side of the device makes contact, should short homing pin to GND
• PCB should be KEYED
  • Only 1 way to put assemble this part.
• Thick Ring Traces
  • Helps to reduce tolerance issues
  • Reduces contact misses
  • Reduces chance of shorting against other contacts
• Multiple pogo per track
  • Contact redundancy
  • Current throughput (not that big an issue)
• Use Easy to Interface Connector
  • Phoenix connector

KEY and Connector Position Concept

See CAD Here

[QuanLe8191]

Slip ring redesign with the dimension fix. I have to increase the outer diameter to 112mm to fix all the traces; also some of the traces widths was changed to fix everything in:

• GND and VCC traces changed from 3mm to 2mm.
• CANH and CANL changed from 2mm to 1.5mm
  Front side:
  

Back side:

3D model of the slip ring (trace side):
OBJ_Drive slip ring design_home pad(trace)_2021-01-15 (1).zip

[QuanLe8191]

3D file of the slip ring (pogo side):
OBJ_Drive slip ring design_home pad(pogo)_2021-01-17.zip

3D file of the slip ring (trace side):
OBJ_Drive slip ring design_home pad(trace)_2021-01-17.zip

[QuanLe8191]

DXF file:
slip_ring.zip
Dimension and Price:

• Pogo:
  

• Trace:
  

[kammce]

1. These traces look way too close to each other. Remove the ring silk screen and space them out further. I took a look at your PCB and it does look like there is space to expand the rings a bit more.
2. The connector for the "Trace" and "pogo" PCBS are on the wrong way. It should be facing away from the traces not towards it as indicated by the silk screen.
3. The connector should be moved more towards the outer edge of the PCB so its easier to connect to by hand.
4. Can you make the part of the PCB that has the connector closer match what is on the concept design I put in such that we do not have to worry about inner corners when it comes to the PCB being manufactured.
5. Can you put the HOME pad at the vertical center of the PCB so it isn't off center
6. The mounting holes look all over the place (I do understand that the lower right hole is meant to be out to place but the other holes are also not nicely aligned either, which they should be to make it easier on mechanical)
7. Make sure that the board is actually symmetrical with it dimensions. Use the smart dimensions tool to make sure that everything is centered in the right place.
8. CRITICAL: The HOME pin is directly connected to the ground pin on the traces side. This will ensure that the HOME pin is always connected to ground regardless of contact with the pogo pin. This must be fixed.

@QuanLe8191

Also the price reduction looks great though! $38 for 10 PCBs (~$76 for 10 Slip rings) sounds like great price for this.

[QuanLe8191]

Make some changes to the circuit based on Khalil suggestion:

The board have some clearance error:
I8QuotationReport_2hq2ygfp_DFM.pdf

[QuanLe8191]

Pogo side:

DRC check:
I8QuotationReport_e1rxdqqw_DFM.pdf

DXF file for both:
slip_ring.zip
@HaadiElahi

[kammce]

This is looking great @QuanLe8191 ! Can you make sure that the hole dimensions are correct and make sure we have enough space between the holes and edge of the board. These could break if they are too thin.

[kammce]

Oh!

1. Also I noticed that the pogo board still have the incorrect orientation of the connector.
2. It would look nice if we had a vertical line that acted as a divider between the PCB traces/pogo pin square area and the connector area.

[QuanLe8191]

This is looking great @QuanLe8191 ! Can you make sure that the hole dimensions are correct and make sure we have enough space between the holes and edge of the board. These could break if they are too thin.

You mean the drill hole at the 4 corners?

[kammce]

One last thing. This may not matter that much, but each ring is a loop. Loops can allow current to flow through them, generated by EMF waves or by magnetic waves, generating spikes and distortions in the signals. It may make sense to break the loop for all connections just as you have done with the GND loop segment. @QuanLe8191

[HaadiElahi]

@QuanLe8191 I modeled the slip ring in CAD and it looks like the connector needs to be moved out so it doesn't interfere with the plate. You can reference the dimensions in the last picture. Basically the connector or trapezoid portion of the ring needs to be extended 4 mm (I changed it from 11.5 to 15.5 mm)

[kammce]

Thanks @HaadiElahi for putting those dimensions together. Makes it very clear what is needed the EE team. This change shouldn't cost too much with regards to the PCB manufacturing.

[QuanLe8191]

I make the change @HaadiElahi
Slip_ring_vA.1.zip

[QuanLe8191]

The inner hole is not cut out. Need to follow SJSURobotic naming standard for PCB

[kammce]

@QuanLe8191 there are lot more things that need to be changed with the design as I mentioned in slack. Can you list them out in a comment using the check box syntax like this:

• task 1
• task 2
• task 3

[QuanLe8191]

Things to improve for the slip ring:

• Change inner hole to board cut out
• Board naming convention should follow the standard: team name, designer, date, rev
• Open traces to reduce EMI
• Larger Via for more current flow on trace board

[QuanLe8191]

@davidbdias96 Object file of the slip ring:
OBJ_Trace vA.2_2021-02-23.zip
OBJ_Pogo Side_2021-02-23.zip

pogo pin:https://www.amazon.com/Yibuy-Piece-Gold-Plated-Probes/dp/B07FPFBFV3/