by

|   Jun 17, 2009 at  2:11pm

Will an unregulated wall unit like this one (GS76H on http://www.maplin.co.uk/Module.aspx?ModuleNo=13453) be suitable, or must it be regulated?

by

|   Jun 17, 2009 at  8:36pm

An unregulated adaptor is fine, as it has onboard +12v and +5v regulators.

by

|   Nov 30, 2009 at 10:38am

How can this be modified for the .01 Hz?

by

|   Jan 24, 2010 at  9:36am

How can this be modified for the .01 Hz? Did you get an answer to this question ? I would like to know too.

by

|   Mar  1, 2010 at  7:43am

the answer to this depends, in part, on whether or not you want to keep the original range still available. Replacing C8 (220nF) with a larger cap would help achieve the lower frequency but you might have a gap between the lower and upper frequency range if you make C8 too big. You could make C9 (2.2nF) larger as well but would be sacrificing the upper range.

An alternative would be to replace the freq switch with an off board rotary one that allows 3 or more selections instead of the two currently shown. Personally, I'd go with 5 positions (2 low, 2 existing, and 1 high) using caps such as 2200uF, 22uF, 220nF, 2.2nF, and 220pF. Note, the caps are multiples of 100 in values except for the upper one (since the existing board goes to about 240kHz and 1MHz is the stated limit). The high end might be a bit touchy with a smaller cap.

You can also use larger potentiometers to lower the frequency but this will make the frequency more difficult to adjust with precision. To get 0.01Hz, I doubt you could pick a potentiometer with a practical resistance value (would be several Meg I think) so best to stick with modifying the caps.

by

|   Jan 19, 2010 at 11:42am

I'm working on a portable function generator. Would two 9-volt batteries in series be adequate for power?

by

|   Jul 21, 2010 at  5:22pm

As long as you have an 2.1mm adapter and more than 14v, yes

by

|   Feb  8, 2010 at  9:36pm

Just built this kit. Been using Radioshack 273-0332 15V 1000mA adapter as a power supply. Make sure you get the "L" plug to use with the power jack if you do the same.

As built per kit: low freq range was measured at 15Hz to 4.793kHz. high freq range was 996.8Hz to 240.8kHz. Upper end of upper frequency range was very difficult to control with precision. I suspect supplied potentiometers used audio taper instead of linear taper based on in-circuit measurements. Nevertheless, the generator appears to function well.

Might be better to use linear pots or place an additional 1K pot in series with P2 & P3 to provide extra-fine frequency adjustment on high-freq end though that would necessesitate some off-board components. I plan to re-mount this board inside an enclosure and move some of the components off-board anyway.

Optional resistors appear to be parallel with P2 and P3, thus would raise lower frequency limit but not raise upper frequency limit. Better to adjust C8 bigger and C9 smaller if trying to extend frequency range I'd think.

by

|   Feb 15, 2010 at 10:19am

This is a cool little kit. I changed a couple of parts and used the acrylic case the Ipod Shuffles come in as an enclosure.
http://tinyurl.com/ygtqnzk

by

|   May 27, 2010 at  7:58am

How could I modify this to get an output voltage of +-80V?

by

|   May 29, 2010 at  4:22pm

The question I need to ask is: how much current do you need? Knowing your current requirements is crucial. I had done a shaker table amp years ago using a BGW (750) commercial sound amp panel, though that was rated 250W into 8 ohms, so "only" would have had 65 volt rails. A +-80 volt rail (audio power) amp would be 450W / 8 ohms: there are a few in this category. Or consider building your own high voltage amp around the LM4702, which is rated to +-100 volts with the right version. Be sure your heatsinking is adequate, and observe typical audio power amp grouding/feedback rules so you don't end up with a high power RF oscillator -- until it overheats and shorts out.

by

|   May 29, 2010 at  3:48pm

Assembled quick, like the dual frequency control pots for coarse/fine frequency control. Be aware: output has no DC blocking capacitor, so sine/triangle output has a 6 volt DC offset to ground. The ability to include one on the board (even if not supplied) would have been nice. Also, no series resistor in sine/triangle output so potential for chip damage if output shorted.

by

|   May 29, 2010 at  9:02pm

The VGND should be used with the Sin/Tri wave output, this removes the offset. A virtual ground is created with an LM741 opamp chip. A more simple DC blocking capacitor would filter the output (especially the Tri output) depending on load and frequency.

Also, the XR2206 chip has short circuit protection inbuilt, so shorting the output will cause no damage.

Regards,
Nuxie1

by

|   Jun 26, 2010 at  5:07pm

Just built the kit. My version came with a jack for a 5.5mm x 2.5mm plug (not 2.1mm i.d. as in the description). Also, I used the 12v regulated power supply that sparkfun sells (very cheaply) and put a jumper between the input and output of the 7812 (the two outer pins); use an insulated jumper so that you don't cause a short to ground. Depending on what size jack comes with your kit, you might need to change the plug on the transformer.

Measured 158 Hz to 289 Khz on my ancient nixie-tube era HP frequency counter.

My only complaint is that the frequency is difficult to adjust precisely because of the two single-turn pots. Maybe a high-quality ten-turn pot would be better. But as a cheap and dirty signal source for testing amplifier circuits and things like that, for $30 you can't complain.

by

|   Jun 26, 2010 at  5:08pm

Meant to say 15.8 Hz on the low end. Should have previewed, etc.