|
EMC of unshielded CAN bus?
|
View:
New views
8 Messages
—
Rating Filter:
Alert me
|
 |
EMC of unshielded CAN bus?
by Oliver Betz
::
Rate this Message:
Hello All,
has anyone experience with electromagnetic susceptibility and emissions os unshielded CAN bus in machinery using ISO 11898-2 transceivers? Till now, I always recommended shielded cables, but now I got a requirement for simpler connectors without shield connections. Since it's used in a machine (not between machines), I guess it should work. TIA, Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
RE: EMC of unshielded CAN bus?
by Steve Corrigan-2
::
Rate this Message:
Hello Oliver,
The TJA1040 and TJA1050 from NXP were developed specifically for these conditions. Cheers, Steve C. -----Original Message----- From: canlist-owner@... [mailto:canlist-owner@...] On Behalf Of Oliver Betz Sent: Monday, November 02, 2009 4:30 AM To: canlist@... Subject: [CANLIST] EMC of unshielded CAN bus? Hello All, has anyone experience with electromagnetic susceptibility and emissions os unshielded CAN bus in machinery using ISO 11898-2 transceivers? Till now, I always recommended shielded cables, but now I got a requirement for simpler connectors without shield connections. Since it's used in a machine (not between machines), I guess it should work. TIA, Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
Re: EMC of unshielded CAN bus?
by Oliver Betz
::
Rate this Message:
Hello Steve,
> The TJA1040 and TJA1050 from NXP were developed specifically for these > conditions. ...as the "EMC optimized" SN65HVD1050A, I guess. The obvious addition is the Vsplit terminal used to reduce common mode voltage and therefore emissions, and the data sheet also tells me that the outputs are better matched than in the PCA82C25x. Well, likely a device willpass emission tests since automotive emission levels are even lower than industrial in some frequency bands. What I'm missing is data about susceptibility, e.g. the high frequency common mode rejection. It would be great to learn about error rates during a EN61000-4-6 BCI test or the EN 61000-4-4 EFT test. Especially EN61000-4-4 EFT test levels are much higher than in automotive tests. Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
RE: EMC of unshielded CAN bus?
by Steve Corrigan-2
::
Rate this Message:
Hello Oliver,
The HVD1040 and HVD1050 pass all of the automotive immunity and susceptibility electro-magnetic requirements and are being used world-wide. However, IEC-61000-4-4 and -6 have not been an automotive requirement. Only in the past couple of years has a large industrial concern began to require -4-4 testing. I don't believe that any of the transceivers on the market have posted any noteworthy results for these test. TI has been working on a product specifically designed for the robustness of -4-4 testing, but I believe it has quite a way to go before release to the public. (things move so slow there it drove me crazy). Cheers, Steve C. -----Original Message----- From: canlist-owner@... [mailto:canlist-owner@...] On Behalf Of Oliver Betz Sent: Tuesday, November 03, 2009 3:03 AM To: canlist@... Subject: Re: [CANLIST] EMC of unshielded CAN bus? Hello Steve, > The TJA1040 and TJA1050 from NXP were developed specifically for these > conditions. ...as the "EMC optimized" SN65HVD1050A, I guess. The obvious addition is the Vsplit terminal used to reduce common mode voltage and therefore emissions, and the data sheet also tells me that the outputs are better matched than in the PCA82C25x. Well, likely a device willpass emission tests since automotive emission levels are even lower than industrial in some frequency bands. What I'm missing is data about susceptibility, e.g. the high frequency common mode rejection. It would be great to learn about error rates during a EN61000-4-6 BCI test or the EN 61000-4-4 EFT test. Especially EN61000-4-4 EFT test levels are much higher than in automotive tests. Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
Re: EMC of unshielded CAN bus?
by Oliver Betz
::
Rate this Message:
Hello Steve,
> The HVD1040 and HVD1050 pass all of the automotive immunity and > susceptibility electro-magnetic requirements and are being used world-wide. Yes, but "pass" might tolerate detected errors and retransmission of packets, and I use only the physical layer but not the CAN protocol, and I can't tolerate bit errors. I'm still confident, so I will make a test setup with special hardware and do my own immunity tests. Greetings, Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
RE: EMC of unshielded CAN bus?
by Steve Corrigan-2
::
Rate this Message:
Hi Oliver,
In my experience in this area, no bit errors were ever tolerated. None. That was the pass/fail criterion. The automotive people send their modules to conduct in-house testing during design. For your testing, be sure to set up some big ugly relays and fire them off next to your bus lines. I'd like to see what you come up with for your tests. I used to enjoy making up new tests for these devices. Cheers, Steve C. -----Original Message----- From: canlist-owner@... [mailto:canlist-owner@...] On Behalf Of Oliver Betz Sent: Tuesday, November 03, 2009 10:14 AM To: canlist@... Subject: Re: [CANLIST] EMC of unshielded CAN bus? Hello Steve, > The HVD1040 and HVD1050 pass all of the automotive immunity and > susceptibility electro-magnetic requirements and are being used world-wide. Yes, but "pass" might tolerate detected errors and retransmission of packets, and I use only the physical layer but not the CAN protocol, and I can't tolerate bit errors. I'm still confident, so I will make a test setup with special hardware and do my own immunity tests. Greetings, Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
Re: EMC of unshielded CAN bus?
by Oliver Betz
::
Rate this Message:
Hello Steve,
> In my experience in this area, no bit errors were ever tolerated. None. good to know. > That was the pass/fail criterion. The automotive people send their modules > to conduct in-house testing during design. > > For your testing, be sure to set up some big ugly relays and fire them off > next to your bus lines. I prefer our old EFT5 generator with capacitive clamp. I plan to make BCI tests with galvanic coupling, and then I will visit a test house. > I'd like to see what you come up with for your tests. I used to enjoy > making up new tests for these devices. Hmm, as a developer I enjoy more if I get something useful working than doing test setups just for research. And there is always so little time... When / if there is any concrete result, I will report it. Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
|
|
Re: EMC of unshielded CAN bus?
by Oliver Betz
::
Rate this Message:
JFTR:
Using two TJA1050 operating at 1MBit/s over 20m unshielded star quad cable, transmission errors started to happen at more than 2kV EFT in the capacitive clamp (grounded DUT). I could test BCI to only 3V and 15MHz but I expect no problems since the calculated common mode voltage should be well below the allowed limits. So immunity doesn't seem to be an issue. Oliver -- Oliver Betz, Muenchen -- Archives and useful links: http://groups.yahoo.com/group/CANbus Subscribe and unsubscribe at www.vector-informatik.com/canlist/ Report any problems to <canlist-owner@...> |
| Free embeddable forum powered by Nabble | Forum Help |
