1
\$\begingroup\$

Hello I have a question that hopefully I can clearly state my question well enough.

I work in television broadcasting and simply take off the shelf converters that take various signals and convert them to fiber. The issue I found was with taking a analog genlock signal such as blackburst or trilevel converting it to fiber and patching it back to coax for the camera (so a transmitter and receiver configuration).

The Sony HDC-P1 camera has no issue with the signal. However, the Sony HDC-P1 "R" series will not accept the genlock from the receiver. The converter is a VOS-1000FT/R, and I found it simply takes the analog video signal converts it to digital via a ADC, goes into a FPGA, gets serialized and passed to a SFP where the receiver de serializes and goes to a DAC, and finally through a filter and to the camera genlock input.

Now the signal looks darn fantastic on the output, no noise in the output and looks to be coupled via a capacitor, BUT has a horrible DC offset of 1.7VDC. Only if I take the signal out of the DAC before the filter (Voltage at .5VDC) and pass it though a RC circuit of 100 Ohm and 100 pF and plug it into the camera genlock port does the voltage drop to .1VDC and it's happy all day.

I am familiar with using opamps and other feedback methods to clamp the signal and the more expensive yellobrik converters and bluebell do a better job, by generating their own analog video signal on the output and clamping all in a isolated fashion. Voltage measurements on both devices and found to be close to zero.

Finally, my question is about designing a product that needs to meet a standard for all devices that it will interface with. How do you determine this and build it in to the product?

The composite signal coming out of the cheap converter looked fantastic and very clean but had the DC issue that the other camera had no issue with. And all the chips are commonly used and look the be implemented using best practice.

How does a large company such as AJA with certainty know there products will interface appropriately with every other device produced is it in the test equipment used?

Thank you very much for taking the time to read this and answer. I'm only 27 years old and can only lean on the backs of a few of my elder friends who worked for Motorola in the 70s and 80s, and these technologies are too far outside of their experience.

\$\endgroup\$
  • 6
    \$\begingroup\$ Interface specifications are the trick. \$\endgroup\$ – analogsystemsrf Mar 15 at 2:52
  • 6
    \$\begingroup\$ That’s where standards (formal and de facto) come in play. \$\endgroup\$ – Edgar Brown Mar 15 at 3:40
  • 1
    \$\begingroup\$ Meeting a standard is one thing. The next problem is when in the field you encounter some other device that doesn't meet the standard it claims to. Bigger products will have bigger test suites in an attempt to deal with this. Microsoft famously had a cart of USB devices when they were first testing their support for it - they wheeled it around to crash computers and leave the developer to address the bug. \$\endgroup\$ – pjc50 Mar 15 at 13:32

Your Answer

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

Browse other questions tagged or ask your own question.