Agilent 33250A Waveform Discontinuity

[dfox416]

I recently acquired an Agilent 33250A function generator for free, since it has an issue. There is a strange jump in the waveform, like the bottom peak is chopped off and shifted upwards (shown in the attached images). The voltage at which this discontinuity occurs scales with the amplitude of the waveform (-250mV for a 1Vpp wave, -500mV for a 2Vpp wave) and the magnitude of the voltage jump at the discontinuity also scales with the waveform amplitude (60mV for a 1Vpp wave, 120mV for a 2Vpp wave). I can see this issue clearly for the sine, ramp, and arb waveforms, and assume that it is also occurring on the square/pulse but can't be seen because of how fast the edges are. Does anyone have an idea of what could cause this, or where would be a good place to start looking for a cause? I appreciate your help.

[Renate]

It looks like your DAC is screwed.
In olden times the MSBs on some DACs would have trims.
But this ain't that.
If it were there would be 3 breaks at 1/4, 1/2, 3/4.

[HighVoltage]

First, check all power rails to be sure they are OK.

But I agree, this is most likely a DAC issue.

[Renate]

Yeah, I thought about the negative rail.
But the signal is not crunched or non-linear and more telling, the entrance and the exit of the offset is identical on the sine.

Well, if thus does arbitrary, you've got overlap.
You can design a custom sine that has that jump inside it so that the output is clean.