Dear all, as a HW designer I am familiar with small DC drives design. Latest project involves a 15A drive at 36V, both DC and BLDC. Can you guys advise me some good book or other source with guidelines for meeting medical level EMC regulations? Thanks in advance!

If you are using in the US then you need to meet the FCC spec, and that is basically shelding the controller and the wires. If you are going to Europe then IEC68000 is a good place to start.

Thank you for your reply. Markets will be Europe and US both. Finding the levels will not be the issue, but how to reach them. Shielding is a good start, but not needing it would be nicer. What filtering options to think about, groundplane usage, while PWM driven, what SW actions to think about. More hints are still welcome!

There is the filter and then there are other things. Yes, ground plane but bonding is important things like this:
- Shielded motor cable - preferable 360 bond at both ends.
- A separate return wire for the motor frame to the drive frame.
- X Caps on the DC bus to the frame
Most EMI problems that you will have will be the clock and the PWM harmonics. The clock is straight forward. Watch out for any wires inside the enclosure - they will pick up the high frequency. Lastly, what do you enclose the drive in? I have had to use 3% stainless fill plastics. Edges where metal connects to metal must be conductive and there must be connections close together.
In my experience - the filter is the obvious and straight forward part. The devil lurks in the not so obvious leaks.

details man, more details needed...
is your 36V used for something else? or just for motor?
you will have to calculate your bus ripple, and then your filter (if you need one).
First thing - do grounding diagram and hire person who can do EMI, not just "EMI BSer". (loud barking dog doesn't bite)

Bill is right in few places, I do not agree with X-caps (X or Y?)....
what if "frame = DC bus" ?
"the filter is the obvious and straight forward part" ??? what if there will be NO filter ?
Motor driver can be on 2x4 inch board, EMI filter for this to meet full 461 - man!, 6x6 inch...
Bill - do you connect shield on both ends if you have 4th "return" wire?

don't just do "GND planes" - may do more BAD than GOOD - think !!!

Google SS modulation- may help by cheating spectrum analyzers.
Do other things with PWM for motor phases...will save you a lot on filter (if you need one).
Don't set PWM freq to anything you like, consult with your motor man - he will tell you what works for his motor (if he knows motors)

if motor frame is not conected to driver frame/bus - can save 6dB by doing very simple trick - can tell you in email, no public.

can do few more things if you don't want 5kG of EMI filter....

Thank you guys! I see some usefull remarks, and find out I have some work to do ;-) More details, well of course it's all topsecret stuf, but will give it a try. 36V source will be powered from a batterypack nearby ( < 1 meter ) and the hardware that is connected to this source will be part of the design, I guess that will give some release? My main worries are on the motoroutputs, length will be < 1m too. As I said preferable without shielding! I try to oversee the consequences, but find it hard get good impression right now. Housing is not a passed station yet, but goal will be plastics, then stainless fill might be good option when encountering problems. 6dB, sounds good... Adam, please see your email. Please let me know what more details you would need to give me a better view!
With regards Rudi