[quote=“VEL255”]Nice! There seems to be added a true RMS value display and possibility to enter the voltage as true RMS too. This is very useful feature.
- I immediately used it to check my (newly purchased) multimeter’s RMS accuracy with different waveforms.
The possibility to enter fine tuning multipliers is nice feature too to “calibrate” the generator.
Maybe the defult setting should be 1.00000 and 1.000
There seems to be a minor “international” problem in this kind of “comma separator county” - doesn’t accept my figures entered…
It seems that the “Library” button is missing from this version…[/quote]
I have corrected the entry and display ussues with the fine tuning parameters–being a newcomer to creating “international” applications there has been a bit of a learning curve here. It turns out that the VB6 Format() function does not recognise the regional setings and insist’s that the format string parameter use periods. I am using that function to display the values in the LostFocus event and in the Current Settings frame title. The values do default to 1.00 however the Format() function was really messing them up.
I apologise for forgetting to mention that in order to reclaim some space I changed the Load Library function from a command button to right-clicking on the libary filename. Right-click on the name and the open dialog will pop-up.
You will also note that when running a library function the “pk-pk” and RMS labels will change colour (to maroon). Hovering over those will display the tool-tip “Pk-Pk and RMS are waveform dependent”. Since the libraries do contain the actual point-by-point defintion of the waveform it should be possible to use that information and the amplitude and offset values to calculate the true RMS of a library waveform–if I get ambitious maybe I’ll work on that.
A clarification: The RMS entry is made in “centered around 0.0V” values, I.e. the entered value does not provide for the effect of DC offset on the true RMS value. As the parameter sent to FGULink.dll is the peak-peak value any value entered into the RMS settings is simply corrected to the peak-peak value for the selected waveform (sine, square, and triangle) and that is used to control the generator. The “true RMS” value displayed a tool-tip when hovering on the labels is calculated usng both the peak-peak and offset settings as shown below
true RMS = (((pk-pk * kRMS)^2) + (Voffset^2))^0.5
Where kRMS is a constant related to the waveform (0.35355 for sine, 0.5 for square, and 0.28867 for triangle waves), this is the square root of the sum of the squares of the waveform RMS and the DC offset.
Rather than add more screenshots to this thread I have edited the message above to shown the latest revisions to the forms.
I also reverted to the 4 significant digit displays for the voltages. The 4th digit may not do anything but IMHO it looks nicer.
An added option is the ability to hide the FGU.exe application–though I sort of prefer to see it as a validation that things are working as intended.
Regarding the frequency resolution and accuracy, I continue to be very impresed with this instrument. I got together with the local shop I spoke of on Monday and we once again calibrated my HP 5316A counter to his rubidium standard. I have a custom oven heated (using a PTC thermistor) 10 MHz timebase in 5316A that is very stable once up to temperature.
Using the 1.000015 calibration setting shown in the modified message above, I programmed the PCGU1000 to output a 1.000015 kHz sinewave, here is a link to a .avi video of the readout on my HP 5316A counter.
You cannot beat that, I watched it for over 30 minutes and it was nearly rock stable at 1.0000150 kHz with and occasional blip to 1.0000151.
Here’s a table showing the results (also to 8 significant digits) of setting the PCGU1000 frequency to 7 significant digits, from 1000.001 to 1000.005 Hz, and the HP counter’s readouts at 8 digits.
These are the real numbers, with no manipulation of any sort–very impressive for a $200 instrument!
So it is possible to send the dll a frequency with 1.4 µHz step definition and effectively get the corresponding frequency (taken apart the master-clock precision/stability)?