SDG2122X only 1 Mhz with triangular wave ?

[ebclr]

With sinewave everything just nice at 120 Mhz.

But the maximum frequency that I can get with a triangular wave is 1 MHZ, Is that correct?

[WackyGerman]

Yes , just look at the datasheet 

[tautech]

Arb max is listed as 6 MHz so you might be able to get that via some trickery.
IIRC rf-loop could coax 50 MHz out of an SDG1000 unit with the SCPI command set.

[WackyGerman]

You re right tautech . Didn t thought for the trick . Checked it with my SDG 1010 . Generated an Arb - wave with Easy Wave and sent it to my generator . Works really well and I was able to raise the frequency to 5 Mhz . 

[ebclr]

Why they make this?

Apparently no reason to limit to 1Mhz

[Kleinstein]

With the waveform generation via memory and DAC, it is normal to have a considerable lower limit for the triangle waveform. The sine waveform using DDS can go up to the point of about 2.5 to 3 sampling point per period - this is because the reconstruction filter can be really good in restoring a sine from those few points. However this does not work that well with a triangle or arb waveform. So at higher frequencies the waveform gets more or less distorted. Here the limit is slightly arbitrary, as it is a slowly starting decline of quality when the frequency is increased to much. For a triangle something like 100-150 sampling point might be a reasonable limit - so maybe 1/50 of the maximum sine frequency.

With the arb waveform there might be a chance to use a little of "pre distortion" to compensate for the onset of the filter. So using the arb way might be able to get a slightly high frequency triangle (at the cost of processing needed) and maybe less fine frequency steps. For the 5 MHz shown one can already see an not so ideal shape.

[WackyGerman]

Didn t do the job really well when I measured the triangle wave . Used a 1m cheap bnc cable and didn t set the generator and the oscilloscope to 50 Ohms . Now the triangle wave looks better . Still not perfect , it s a little distorted 

[DaJMasta]

You can get something close to a triangle wave up to about 15MHz on my SDG2082X, generate a sine wave, turn on harmonics, set about:
5%Vpp amplitude second harmonic with 270 degree phase
10%Vpp third harmonic with 180 degree phase
2.5%Vpp fifth harmonic

You really need the 5th harmonic for it to look right, so with your 120MHz sine bandwidth, your cutoff for this approximation would be 24MHz.  I guess in theory I should be able to get a 16MHz one out of mine.


With a little tweaking it could probably look nicer, but aside from a tiny bit of wobble in the ramp and a rounded peak, it's serviceable.

[mmagin]

This is not unusual among cheap function generators.  If I recall correctly, even some of the lower-end old ones from HP (30-35 years ago) had such a limitation.

[Kleinstein]

The old, analog (and maybe with PLL stabilization) generators also might have a lower limit for the triangle. But this is due to the output amplifier, so maybe at 1/5 - 1/10 the sine frequency. The DDS based generators usually have a lower limit at maybe 1/20 - 1/50 the maximum sine frequency.

[ebclr]

We are talking about 1/120 here