• My problem I am experiencing is with the circuitry or IC combination to follow. I don't need a answer of what exact IC models I need, as that would be plain stupid. What I asked originally about the op-amp was because I believe it was the correct thing to use if my signal that is induced does not inherit the same peaks of voltage in the original wave form.

  • My hypothesis of how the circuit should work is that when the signal from CAN_H or CAN_L is induced into sperate strips -> the signal is then rectified to the correct wave form depending on what gets distorted then -> using an IC (comparator as suggested by Andy, also after examining graphs an other data ) to ensure the correct digital form (if an analog form generates in the previous stage) and then sending it to an external monitoring system. All without physical obstruction of any lines. Is this thinking correct.

  • What I want from the circuit is that when I connect it over CAN lines using small copper pads on a PCB it then needs to send this (image) out after reading to an external monitoring system .
    CAN BUS lines

  • I cannot use the standard methods of connecting to the J1939 connector on the heavy duty vehicles as it becomes an issue of space and tampering/fiddling by drivers.

My issue only lies with the overall circuit choice. As Andy aka has pointed out what he has done and that this can be done. I only need to be able to reproduce those signals. Hence the link to this.

I am sorry to have explained my question vaguely, it generally sounds better in ones' head before they speak or write.

  • \$\begingroup\$ Try writing your question in English, paying attention to what is the subject and what is the verb. It seems you have a question about interpreting CAN signals, but beyond that it's hard to tell what you are asking. \$\endgroup\$ Jul 8, 2014 at 17:36
  • \$\begingroup\$ HI, thank you for the comment. basically I want to read the CAN data through induced signals and thus relay it to the other modules. My problem is would I be able to use the op-amp for this. And I will edit it. \$\endgroup\$
    – RSM
    Jul 8, 2014 at 17:39
  • \$\begingroup\$ "Induced" how? If you don't want to connect to the CAN lines directly, then the big question is how exactly you expect to pick up the signal, not what opamp you might use afterwards to turn it into a digital signal. \$\endgroup\$ Jul 8, 2014 at 17:44
  • \$\begingroup\$ Considering the fact that the CAN bus line are in a twisted pair to reduce the coupling effect between them and surrounding cables due to the rate at which they operate in data transmission. My thought would be to use etched plates on a PCB placed over a section in turn linked to the opamp. As explained in the product linked in the question. \$\endgroup\$
    – RSM
    Jul 8, 2014 at 17:47
  • 1
    \$\begingroup\$ @HannoBinder tapping in causes issues with the people who service the vehicles and the company doesn't want to be held liable. And I would probably use the circuit for each line. \$\endgroup\$
    – RSM
    Jul 9, 2014 at 12:15

1 Answer 1


I've done it (not on CAN) like this: -

enter image description here

The top section shows some data and what you will receive (shown just below) will be a highly differentiated form - just the edge information comes thru - positive edges on the data result in a positive spike and negative data edges result in a negative spike.

Cutting a long story short, you apply upper and lower detection thresholds (a comparator with hysteresis for example but there are other ways) and what the comparator produces is an exact repro of the original data.

Things to watch out for: -

  1. Use a small pick-up loop so that you are not taking significant energy from the signal - the loop I used was about 25mm long by 6 mm wide
  2. Feed it into a low noise fast op-amp circuit - I had a transistor circuit that was phantom powered on mine because it had to sit 10m from the rest of the electronics
  3. Setting the comparator threshold/hysteresis isn't too bad but will take a little bit of setting up - I used a circuit that automatically tracked amplitude and adjusted the thresholds automatically but the "sniffer" I used was sitting in the presence of a large 600 kHz H filed of about 1000 at/m and this had to be contended with.

Mine runs at 80 Mb per second but I think yours will be substantially less so you might need more gain because induction (is better at higher frequencies.

  • \$\begingroup\$ The comparator wave form, is that a digital form, I ask as all I need to do is reproduce the CAN signals wave form after coupling. \$\endgroup\$
    – RSM
    Jul 8, 2014 at 22:22
  • \$\begingroup\$ The output from a comparator is usually digital in nature. \$\endgroup\$
    – Andy aka
    Jul 9, 2014 at 7:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.