Oscilloscope Aarontek DSO338

DSC_6103

This is a small oscilloscope with the most common functions.

Official specifications:
DSC_6078

It arrives in a brown cardboard box.

DSC_6098

Inside is the oscilloscope, a probe, a usb charging cable and a instruction sheet.

DSC_6112

Lets first take a look at the probe.

DSC_6114

It has the usual 1x and 10x setting and even a color coding ring for multichannel scopes, it is not much use here.

DSC_6115
DSC_6116
DSC_6117

The tip is fairly standard for oscilloscope probes.

DSC_6113

With a ground lead with alligator clip

DSC_6111

What is non standard is the connection to the oscilloscope, it is with a mini jack plug, not BNC.

DSC_6104 DSC_6109

DSC_6106

The oscilloscope is a box with screen and keys on the top side.

DSC_6107

The mini jack input is at one end.

DSC_6108

On-off and USB charging connector is at the other end.

DSC_6110

All user interface is with a couple of buttons:
DSC_6260

DSC_6300

Welcome screen.

DSC_6301

This is daylight mode screen.

DSC_6303

A look at saved waveforms, it is possible to use the arrow keys to move around and AC/DC, MENU and RUN/STOP is "soft keys", i.e. the functions listed at the bottom of the display. 1X/10X and 50% are delete keys.

DSC_6302

Viewing a saved waveform, the time and voltage setting is missing from this display. The VPP and F values will be saved with the curve, but not the measurement popup. Use arrows to select next.



Testing
Voltage

A check of the volt ranges with DC voltage, they are not precise at the lower values, but good enough for most work.

DSC_6258 DSC_6259

A 1kHz 100mV rms input signal, it looks fine except the RMS calculation cannot handle a square wave.

DSC_6261

Duty cycle measurement works very fine, here I use a 2% duty cycle at 1kHz and the scope shows it nicely.

DSC_6262 DSC_6263

100kHz also looks good, but there is some spikes on the square waves, I will look at them below.

DSC_6264 DSC_6265

1MHz also looks good, but at 3MHz the spiking is a bit much.

DSC_6267

My connection to the function generator was done this way with a termination, but the probe ground lead may be a problem.

DSC_6269

Lets try this instead with very short ground connection (Some probes includes accessories for this type of connections).

DSC_6268 DSC_6272

The spike is still present, but maybe a bit smaller.

DSC_6275

Lets switch the probe to 10X range without telling the scope about it (To avoid it selecting the high ranges). This reduces the spikes a bit, but voltage calibration is not 1/10 (Probes usual has much better frequency response in 10X range).

DSC_6276

The manual trigger is always at the left edge of the screen, but level can be adjusted.

DSC_6278

Here I try to do a "real" measurement, i.e. checking timing on some neopixel data. This is very fast pulses and the high time must be within some limits to work. I believe the scope is good enough to show it, but not without trouble. The trigger do not fire every time, I had to send the data many times before I got a capture.

neopixel

My usual (and much more expensive scope) captured on first try and the higher resolution is a advantage.

DSC_6279 DSC_6280

A look at a servo signal, the pulse is 1.5ms long, there is some timing error on the scope. Slowing the timing down to get more pulses and adjusting the pulse with to 2ms also has some timing error, there is exactly 20ms from start of first pulse to start of next pulse.

DSC_6281

Switching to auto trigger and using the measure menu shows the correct values.

DSC_6282

A single character at 9600 baud, it is a "z" with code startbit-01011110-stopbit (lsb is transmitted first). It cannot decode the data or collect more than one screen at a time.

Aarontek%20DSO338%20Charge%20curve

A full charge of the build in battery from USB, it looks like the rated 3000mAh is correct. With close to 8 hours charge time it is not fast to charge, but then the runtime on a full battery is also fairly long with an estimated 13 hours.



Tear-down

DSC_6284

It is easy to open with four screws on the back.

DSC_6285

The battery is fairly large and unmarked (I also checked the other side).

DSC_6286

Two more screws and the circuit board was loose.

DSC_6287

The front has holes for each button, with a membrane on the front.

DSC_6288

Here are all the buttons and the good looking display.

DSC_6290

DSC_6291 DSC_6293

DSC_6292

DSC_6298

There is nothing hidden behind the display.

DSC_6289

The main processor is a STM32F407VET6, this is a ARM Cortex M4. This processor has build in ADC, but for this application an external one is used (AD9288BSTZ-40 dual 8 bit 40MSPS ADC), it looks like both channels are connected to the processor. It do also have a 64Mbit memory to store waveforms in (25Q64JVSIQ)
The input range is select with a pile of optocouplers (PC817) and resistors bypassed with capacitors. For DC/AC selection is used a COC10008N opto mos relay. Between the input and the ADC is a amplifier (OPA356 marked OAAI) video OpAmp.
Behind the USB connector is a TP4056 LiIon charger and behind the switch a 3.3V regulator (TLV70033DD marked ODN).

DSC_6294

DSC_6295 DSC_6297

DSC_6296



Conclusion

With a single trigger setting this scope is much more versatile than the cheaper models, but it do have some limitations as described above. This means when working with fast stuff or want to see precise timing the scope is not ideal, but will often be usable.
I would have preferred a BNC connector on it, but it is not very hard to make my own cable with a mini jack.

The scope can help with many problems, but the specifications do not match completely.



Notes

DSC_6299

This homemade cable could be very useful, if you are using Arduino or other small micro controllers.

The oscilloscope was supplied by Banggood for review.