General
Overall, the DS215H provides an impressive set of features in a very small package and a very reasonable price. After a week of occasional use I’ve become familiar with it’s operation, but haven’t yet used it for any serious measurement or debugging. It appears to be a good piece of equipment that I would likely recommend to others, except for the phantom pulse problem described below. I consider this problem almost worth returning the product for, unless it can be fixed with a software update. As received, it is running V0.0.3.
The signal generator output switching bug described below is a minor annoyance, but should be fixed.
User Interface
The user interface is somewhat awkward at first, though with some experience it becomes reasonably easy to use. Given the physical limitations of such a small device it’s hard to say what might be a better design.
Here are a few suggestions on the English language annotations.
• “Awaiting auto-correction” would be better phrased “Auto-calibration in progress”. Similarly, “Awaiting Auto Measurment” would be better as “Auto-measurement in Progress”. The word “awaiting” in this case leaves one wondering if one is waiting for the measurement to start (perhaps waiting for a trigger event) or waiting for it to finish (which seems to be the intended interpretation). Saying it is in progress makes it much clearer.
• The term “afterglow” is used for the screen trace duration. Although this is perfectly logical, the term that has been in use for many, many decades in English with regard to oscilloscopes is “persistence”. This comes from the days of phosphors in CRTs, where “persistence” was one of the important characteristics of a phosphor. The usage of “afterglow” is slightly confusing to a native English speaking engineer familiar with oscilloscopes.
Oscilloscope
In certain conditions, phantom pulses appear on Channel 1. For example, starting an auto-measurement with nothing connected to the inputs will show a burst of pulses. When the measurement completes these are gone, but moving the Ch. 1 baseline will cause them to reappear. If the baseline is set at two divisions above or below the center of the screen, the pulses are two divisions high. If the baseline is set at the center, the pulses are four divisions high. At other positions of the baseline there are no pulses.
They appear as positive-going 20ns pulses with sporadic intervals between them, sometimes as short as 20ns, often longer. They are mostly apparent when the time base is set between 20ns/div and 200ns/div, though they also show up at a setting of 10us/div. Changing the Ch. 1 volts/div has no effect on the pulse height.
The trigger circuit responds to these pulses as though they were real inputs. Turning Ch. 2 off or on has no obvious effect on the Ch. 1 phantom pulses, and no phantom pulses are observed on Ch. 2. If there is an input signal the pulses appear superimposed on that signal. With no input signal, it makes no difference whether the input is shorted or open.
The behavior with regard to these pulses is actually more complicated and therefore difficult to describe in words. A video can be provided if there is interest.
Signal Generator
• The On/Off indicator says NO instead of ON.
• Changing the On/Off state changes the indicator but doesn’t actually affect the output until the frequency setting is changed. It took a while to figure out why I wasn’t getting any output even though the green indicator was lit. This appears to be a software bug.
• Output waveforms are clipped about 10mV above ground. This is very noticable with the output set to any of the continuous functions (sine, sinc, etc.) at 0.1V amplitude.
Screen/Waveform Capture
When you do a screen capture, in addition to placing a .bmp file in the “Screenshot file” directory, there is also a corresponding .bin file placed in the “Screenshot simple file” director. This appears to be a series of 8-bit numbers representing the actual trace values, but I can find no documentation to confirm this. This could be a useful feature, and although it seems simple to reverse engineer it, if it is an undocumented feature it is unwise to rely on it as it could change without notice.