2
\$\begingroup\$

I've got an old-school HP 8656B Signal Generator, for a 433MHz RF project.

Typically, I've used the internal 400Hz or 1kHz modulation but for the receiver in question I need a signal source that is more like 100Hz.

What would you use to generate a 100Hz external signal for my signal generator?

I've looked through the manual, and they just reference using an external source without specifying what exactly you would use.

Would you use just a generic signal generator like the Siglent Technologies SDG805, and you pulse the RF signal generator with a square wave?

Thanks, -Tim

\$\endgroup\$
  • 1
    \$\begingroup\$ Rectified 50 Hz mains, Arduino sketch, amplified Android tone generator, PC tone generator, 555 timer, bass guitar, music synthesiser, ... \$\endgroup\$ – Transistor Jun 28 '17 at 22:33
  • \$\begingroup\$ So... I think the target is to have something with BNC connectors... \$\endgroup\$ – Leroy105 Jun 28 '17 at 22:37
  • 1
    \$\begingroup\$ you MUST define purpose and specs like Voltage, current, waveshape if it matters. or Power, Load f tolerance and spectrum etc. Your PC audio output is handy .. Use Audacity for a sine out. BTW RF is wireless not wired and 100Hz is not RF \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jun 28 '17 at 23:08
2
\$\begingroup\$

If you have got that HP 8656B, you shouldn't have any trouble to get a proper low freq signal generator to be connected to the mod input of HP. SDG805 is more than a valid one, if you can get it. Only remember to have its output amplitude so low that HP keeps all smoke inside. I would add an attenuator between them for safety.

Modulating with square wave sounds suspect. You should be well aware of actual available modulation specs in HP. Modulation with square wave in theory demands infinite bandwidth. Hopefully in your team exists one that understands the modulation mathematics.

User @Chris Stratton has inserted the following comment:

Most uses for 433 MHz radios would expect OOK (=on-off keying) or FSK (=frequency shift keying) modulation by a square data signal. However, the normally utilized transmitters may have a narrower modulation bandwidth than the signal generator does (in order to avoid splatter), so it might take a profiled modulating waveform to accurately simulate with a signal generator the kinds of transmissions likely to be encountered

Thanks to Chris Stratton!

\$\endgroup\$
  • \$\begingroup\$ Great answer, thanks! I'm trying to keep the smoke in the HP too, so not just hooking up it to anything that generates a signal... it is kind of weird the HP manual doesn't really. The HP 8656B manual specifies maximum voltages, not so much what sort of signal you put in though... Typically, I just have used the internal modulator at 400Hz for firmware testing -- but I need to slow it down. I'll do some digging on the actual modulation when I feed it an external signal. \$\endgroup\$ – Leroy105 Jun 29 '17 at 1:10
  • 2
    \$\begingroup\$ Most uses for 433 MHz radios would expect OOK or FSK modulation by a square data signal. However, the normally utilized transmitters may have a narrower modulation bandwidth than the signal generator does (in order to avoid splatter), so it might take a profiled modulating waveform to accurately simulate with a signal generator the kinds of transmissions likely to be encountered. \$\endgroup\$ – Chris Stratton Jun 29 '17 at 5:03
  • \$\begingroup\$ In my case, yes it is an FM Modulation (ie. FSK), which is why I was thinking of a square wave as the input (it's really the period I want to change, the signal generator will still generate the desired FSK deviation). I've got a REALLY CHEAP $30 USB signal generator coming and some BNC connectors. I'll have to see what type of signal (amplitude namely) and see if it needs attenuation. \$\endgroup\$ – Leroy105 Jun 29 '17 at 17:43
  • \$\begingroup\$ @Leroy105 Probably you can set the right peak deviation. But to have a realistic signal, the frequency changing speed must also be right. This is a problem if the generator couldn't change its frequency fast enough. As Chris Stratton told, you probably must make the change slower (=program a ramp to the square wave or insert a low pass filter to the outpit of your lo freg generator). This is where the modulation math capable member is needed. A spectrum analyzer probably also is available, so the right signal shouldn't be a problem if the requirements are known. \$\endgroup\$ – user287001 Jun 29 '17 at 17:59
  • \$\begingroup\$ I'm guessing the signal generator is going to work okay -- I mean the signal is not fast. Yes, I've got a spectrum analyzer as well... It would be quite hard to do the RF work without both. I wouldn't know where to start on the math, I've just got the tools to see the output on my spectrum analyzer and then the target board as it demodulates the signal. Basically, I can fine tune the signal generator to get the output on the receiver. Now can I get it fine tuned well enough, with a toy grade usb signal generator -- to be seen... I'm not with familiar with a signal generator. \$\endgroup\$ – Leroy105 Jun 29 '17 at 18:26

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.