Function generator with sweep function

[WaveyDipole]

Originally Posted by MrBungle

If you add a pot (or any variable voltage source) to the design which you can switch in instead of the ramp generator as an option, that can be used to indicate the centre frequency. Connect a counter to the TTL output on the generator and move the pot until it outputs 455KHz on the counter, then align the scope X position with the centre of the graticule. You can also use that to measure Hz/div etc.

I gave your suggestion a try, but am still not entirely confident that what I am doing is correct. I think it does work better that way though. On the oscillator, there is no switch for ‘VCO mode’ so the course and fine controls always influence the output frequency. I have not found a manual for the sig gen, so I am presuming for the moment that the course and fine controls need to be set at minimum. I was setting the offset and gain on the sweep gen to zero and using the sig gen controls to set the centre frequency (because the fine control makes it easier) and then turning the gain up as needed on the sweep gen. Using the offset control on the sweep gen it is very difficult to get the frequency even close to what it should be, so I might have to add a pot for a ‘fine’ control.

I didn’t think to connect the counter to the TTL output. The oscillator does have a digital readout, but it is driven by analogue controls rather than and encoder or direct entry and the display is off by a few Hz, sometimes a few tens of Hz from the FC , but I trust the FC more than the sig gen display.

Originally Posted by mhennessy

Yes, that's effectively what I do with my Wavetek 162 that drives the RF oscillator. Adjust the sensitivity of the channel driving the X plates to get whatever you need (e.g. 4kHz per horizontal division for an AM set). Switch back to the oscillator when you're done.

I know that the sig gen requires 0-3v to drive the VCO for a full 0 to 2 sweep in a given range, so for example I am using the 0-2MHz range and could probably calculate the start voltage and the ramp height to give say a 405kHz to 505kHz sweep to give 10hKz per div. Not easy to set this up accurately though.

To implement that on this circuit, I reckon you just need to disconnect the 555 from the op-amps (you might need to connect it to a fixed DC voltage from a couple of resistors - perhaps a pair of equal ones to give Vcc/2, given that the ramp goes from 1/3 to 2/3 of Vcc). You have the DC offset control already, so no need to buy another pot and knob. A toggle switch on the front panel would be easy to add. Just a thought

Yes, that is a thought. I have just been turning the gain control to zero, but it would mean less wear on the pot and the control could be left set to where it is needed. I will see whether I can squeeze one in somewhere.

[MrBungle]

Originally Posted by WaveyDipole

I gave your suggestion a try, but am still not entirely confident that what I am doing is correct. I think it does work better that way though. On the oscillator, there is no switch for ‘VCO mode’ so the course and fine controls always influence the output frequency. I have not found a manual for the sig gen, so I am presuming for the moment that the course and fine controls need to be set at minimum. I was setting the offset and gain on the sweep gen to zero and using the sig gen controls to set the centre frequency (because the fine control makes it easier) and then turning the gain up as needed on the sweep gen. Using the offset control on the sweep gen it is very difficult to get the frequency even close to what it should be, so I might have to add a pot for a ‘fine’ control.

AFAIK the VCO inputs are summing with the manual control. I wonder if it's better or easier to apply a fixed voltage to the VCO input which is mid-range in the ramp and then use the manual control to set up the centre frequency.

I didn’t think to connect the counter to the TTL output. The oscillator does have a digital readout, but it is driven by analogue controls rather than and encoder or direct entry and the display is off by a few Hz, sometimes a few tens of Hz from the FC , but I trust the FC more than the sig gen display.

Main reason to do this is so not to load anything with the counter. The TTL output is separately buffered. Generator display is probably quite accurate if it's like my TG215. The precision isn't as good as an external counter though.

I've got a TG215 here and some 455KHz resonators. I might see if I can work something out if I get time.

[WaveyDipole]

Originally Posted by MrBungle

AFAIK the VCO inputs are summing with the manual control. I wonder if it's better or easier to apply a fixed voltage to the VCO input which is mid-range in the ramp and then use the manual control to set up the centre frequency.

Your comment regarding the VCO input being summing gave me an idea which I put to the test this evening. It occurred to me that the start frequency could be set on the signal generator using the finer controls and a ramp of a fixed height added to give a range. If 0-3v gives the full sweep of 0-2MHz, then 0-0.3v should give a sweep of 200kHz (20kHz per div), so a sweep of 100khz (10kHz per div) would require a ramp height of 150mV. To set and monitor the ramp height and ensure that the offset was always at 0v I used an additional BNC cable connected on a T-piece at the VCO input and connected to the 3rd channel of the scope. I had earlier carefully tweaked a spare IF can by peaking the wave visually on the scope while injecting a fixed 455kHz signal, having in mind to use it as a reference. With a start frequency of 400kHz, ramp height of 150mV and ramp width set to exactly 10divs, the peak on the reference can (assuming it had been correctly peaked) should be placed at exactly 5.5divs. It was very close indeed. I repeated the procedure using a ramp height of 0.3v and start frequency of 350kHz, and again it was within a smidgen of where I expected it to be. This seems to be the most accurate method that I have found so far. When time allows I will repeat this experiment for a 10.7MHz FM IF can.

Originally Posted by MrBungle

Main reason to do this is so not to load anything with the counter. The TTL output is separately buffered. Generator display is probably quite accurate if it's like my TG215. The precision isn't as good as an external counter though.

During my experiments this evening I discovered that the readings on the FC were being affected by a loose BNC plug. With the ramp generator disconnected and the plug properly seated on the BNC connector, the sig gen and FC seemed to be within a couple of kHz of each other.

Originally Posted by MrBungle

I've got a TG215 here and some 455KHz resonators. I might see if I can work something out if I get time.

Thanks. It will be interesting to see what you come up with should you get the time.