I received the HDS2102S earlier today.
S/N 2152*** (presumably built in week 52 of 2021)
SW: V1.3.3
Checksum: 67f52158.
I scanned the earlier parts of this thread, and tested most (if not all?) of the reported shortcomings. Since tly posted the nicest list, I'm using their order, even though not all of the bugs were first reported by them. You can find their postings here:
tly post a-etly post f-gtly post h-ia) Fixed. "Channel 2 is Off" message never appeared for me during my usage
b) Fixed. I was able to manipulate the frequency after returning to the AWG screen
c) Fixed. Upon returning to the Osc screen, I see a sine
d) Present. Changing from sample to peak affected the frequency measurement (but not the counter). I looked around the waveform after stopping, and didn't see any double edges, but maybe I just didn't look closely enough.
e) Fixed. There are still a number of quirks in the AWG settings (for instance, the allowed adjustment ranges change when toggling between frequency and period), but this particular one is fixed.
f) Fixed. Recalling cursors from a saved config appears to behave as expected.
g) Fixed. Time cursors remain in units of time.
h) Present. Battery indicator jumps from ~6pixels to "low battery" at around 3.76V (measured both cells). This could be intentional, depending on power supply margins needed.
i) Present. However, as explained in an earlier post, an AC trigger coupling with a DC signal coupling doesn't lend itself easily to a reasonable way to display
The behaviors surrounding channel selection and the auto button,
as described by Flopdoodley are also still present.
The buzzer option is still missing,
as described by rn777Similar to luma, I was excited to see how the unit performed when provided a fast edge input from one of
Leo Bodnar's fast edge generators. Since the single channel sampling rate is 500MHz, the absolute best case would be 2ns. I measured closer to 2.8ns by eye, but as luma mentioned, it's hard to tell with much certainty on the unit itself. Maybe extracting the waveform and analyzing on a computer might provide some insights. I suspect that due to the sin(x)/x interpolation, the real edge time would be best measured by doing some statistical analysis on a bunch of the measured edges.
I investigated the pulse performance based on how it was input. I tried a direct connection (the pulse generator has a BNC connector), using the spring ground on the included OW3100 probe, and also tested the performance of two cheap Daniu P6100 probes, one on each channel, but only one channel at a time. I saved all four of these configs as reference waveforms, then attached the clip ground on the OW3100 probe to the edge generator, to get the screenshot "pulse_inputs_2.png".
I also tested out the 20MHz bandwidth limiter, by capturing a reference waveform of the (directly connected) edge generator, and then turning on the bandwidth limit. This comparison is screenshot "pulse_bw_limit.png".
I popped the back off, and was surprised to discover that the batteries are literally taped into the unit. I don't recall seeing that on earlier models. I means that in order to swap them out for the first time, you have to first remove the whole back cover. I didn't realize this before attempting to pry them out with a plastic spludger, since I didn't want to open it with power still applied, thus the small scuff marks. The board is dated 2021-09-16, and has a revision of HDS2102x_main_V1.0
Home
Help
Search
Login
Register
