eBike Master Switch Design

Doesn't that gap between B- and C- look too narrow?
I mean there might be up to 100V from what I see here.
Unless everything is sealed of course.

Suggestion, widen the gap and use another track inbetween for the gates, getting rid of the resistors and adding more distance in a way.

Looks great BTW.
 
PCB Voltage Breakdown

Thanks for your comments.

The gap is presently 0.020". PCB is said to have 40 volts breakdown per 0.001" spacing, so we should have a breakdown voltage of about 800V here, perhaps more since the board is solder mask coated. I could open it up slightly, but very much would be difficult as the FET leads are already specially bent (and getting long) and we want to keep the leads short for various reasons.

How much spacing is in the motor controller between the pads for the leads of the TO220 FETs?

References

http://www.smps.us/pcbtracespacing.html
 
Was more thinking of air than PCB, but you are right, not only do you have some sealing as mentioned, but the dielectric strength of air is around 1.5KV for the 0.5mm gap (or more with your wider gap now).
Proves again. looks can be deceiving, sorry.
 
Creepage distance:
http://www.ce-mag.com/ce-mag.com/archive/01/03/ProductSafety.html

Depending on material and pollution level, somewhere around 1.0mm for 100v is recommended.
 
Hi Richard,

TO220 pads are normally closer than 1mm, even the leads coming from the part are pretty close together. Has that much space been used on the Zephyr boards, or other commonly used boards? How about in the motor controllers? Have they had problems?

Looking at IR specs, the lead spacing is 2.54mm center to center, and the lead width as it exits the body is 1.78mm max, so the spacing there is as low as 0.76mm. They make parts up to what voltage in that package?

These boards should be coated after construction, certainly if contaminated almost any spacing will be insufficient. Spraying them with Krylon, coating with epoxy or PC board coatings is highly recommended.

I modified the part layouts for the TO220, moving the leads farther out (lengthening the gate and source leads). Spacing is now over 1mm.
 
People use TO-220 parts at 100v all the time. Whatever the standard leg spacing is doesn't seem to be a big problem unless you get it wet.

Conformal coating, potting, etc. pretty eliminate the creepage distance issue and you just go by the dielectric strength of the insulation, which is plenty. Creepage standards assume some board contamination.

Where you get into problems is on sensitive FET gate lines or op-amp inputs. Just breath on them and they can misbehave. I like to imagine the thing getting sprayed with water and see what kind of voltage changes you would get from the resistance of the water path. For example, I usually wire throttle connectors so the ground is in the middle, between the signal and 5v lines. A water path will tend to pull the throttle down. A fail, but fail safer than having the 5v line next to the signal. Sometimes you can place a grounded shield trace on a board to prevent such things.

Anyway, you get my drift. Don't over analyze the thing but at least be thoughtful of potential failure modes.
 
Good to see this project progressing.

Can you suggest a P FET for Q5?
Element14 have a free shipping deal this week , so I'd like to order a few, but I'm not sure what specs to look for.

Thanks.
 
Alan,

I PM'd earlier about my interest in building a test board. In the interest of full disclosure, I am a totally novice electronics builder. I am a molecular biologist, with no electronics background. I don't have the parts required for the board except for the 4110 MOSFETS, which I do have. I have looked over the BOM and determined that the parts can be acquired from Mouser with no difficulty. Even as a novice, the layout looks pretty straight forward, although I not so sure about the connections to J1. Presumably, J1 connects to a SPST switch that makes this a "master switch", but I would need help in understanding the connections that achieve that goal. As I indicated in an earlier PM, I would be happy to pay for the board and shipping.

There may be advantages in involving a relative novice to test the board, as you will be able to see how the clueless can screw up your design. Or it may be too early to test this at this stage. Of course it is your choice.

Rich
 
It will render with components if you have the component models, I haven't looked into that at all.

It will output VRML, but I haven't worked with that or those tools. I would expect there's some way to pull it into Solid Works but have never tried to, and I don't have Solid works, etc myself.

You can download DipTrace for free, so experimenting with it would not be hard. Pull in a sample board design and hit the 3D button. :)
 
Info, BOM, etc for V1.1 boards will be kept in this one post here:

http://endless-sphere.com/forums/viewtopic.php?f=14&t=54225&p=825100#p825100

The first post in this thread has been updated to include links to each of the board revisions. The information about each revision will be updated in those postings.

Some boards are out to testers.
 
You must log in or register to reply here.

Similar threads

X
Converting a dual motor scooter from 2 controllers to 1 temporarily. (kyesang ESCs)
Replies
1
Views
343
chuyskywalker
chuyskywalker
thorlancaster328
Designing the ultimate (400A+) motor controller - what features should it have?
Replies
11
Views
3,045
SammyMc
S
J
Integrated, Scalable Solar/Battery/AC Charging/Inverting/EV DC Charging and More
Replies
1
Views
1,972
borrelan
B
O
EBiCS Firmware on a M365 STM32 Controller
2
Replies
46
Views
9,137
M0g13r
M
Beachcruzer
Craigslist Yescom - Did I score or get screwed?
Replies
13
Views
1,586
999zip999
999zip999
Back
Top