Hantek 1008c - defect or just bad?

[Phil1977]

Hi,

has anyone experience with the cheap Hantek-USB-scopes?

I was tempted by the 1008c because its specified with 8channels and 12 Bit resolution. I often do multichannel measurements, mostly I do not need more than 10kHz of bandwidth, but I need vertical resolution and SNR. I knew I can't expect to much from a €100,- unit that´s designed for automotive workshops - but hey, life is boring if you don't give cheap thrills a chance from time to time.

But anyhow, first tests with the function generator were disappointing. SNR and vertical resolution is not worth the 12-Bit label on the first glance. But what´s really disturbing is the mis-compensation of the analogue inputs.

All 8 inputs have a non-adjustable mis-compensation as shown in the screenshot. Some channels are overcompensated like in the screenshot, others are heavily undercompensated, two of eight are somehow acceptable. The signal from the units own square generator is absolutely okay, like shown in the RIGOL-screenshot.

I´ll return the device anyhow - but do you think I should mark is "defect" or can this be the typical performance?





Just to mention it clearly: The measurements have not been taken with an adjustable 10x-probe. The squarewave was directly fed into the BNC-input. Terminating the BNC with an extra 50-Ohm adapter didn't change anything.

[bobxyz]

I have a 1008 from ~10 years ago.  I bought it for automotive use as a cheap scope that could be accidentally damaged without much, if any, grief on my part.  IIRC, there was a teardown somewhere that showed that channels 1 & 2 had extra circuits & shielding.  I'm not all surprised that it doesn't meet full specs on all channels.

The latest software is much better than the original, but it's still buggy and will occasionally disconnect/reconnect on USB (more frequent when using more than 1 or 2 channels).  It's been a couple years since it BSOD'd on my Win10 laptop, but that used to be common with the early software.

[Phil1977]

Anyhow, every PCs sound input can digitize a 1kHz square wave better than this thing.

Do people in automotive workshops really only want the timing information and signal shape just doesn't matter?

BTW, the shown channel is channel nr. 1. I don't want to open it because I return it to the seller.

[radiolistener]

I often do multichannel measurements, mostly I do not need more than 10kHz of bandwidth, but I need vertical resolution and SNR.

You can use several USB soundcards with 24-bit 192 kHz ADC. It will be much better than this toy.

The only thing that needs to be careful is that using it in automotive can lead to damage, because automotive can have some high voltage pulse. So, there is needs some protection, but I suspect that this huntek also don't have protection on the input.

[darkspr1te]

I also have one of these units, i did get it to function, sort off but gave up as the time invested into getting it all setup each time was annoying so i just reverted back to my rigol ds1054z.
it uses a stm32f103r8 as the mcu so dont expect much.
the signal generator was also poor and i could never get it to reproduce a signal for my diy ecu crank input,
it is supported by sigrok so you might have more luck with that.




darkspr1te

[Phil1977]

I often do multichannel measurements, mostly I do not need more than 10kHz of bandwidth, but I need vertical resolution and SNR.

You can use several USB soundcards with 24-bit 192 kHz ADC. It will be much better than this toy.

The only thing that needs to be careful is that using it in automotive can lead to damage, because automotive can have some high voltage pulse. So, there is needs some protection, but I suspect that this huntek also don't have protection on the input.

I don't want to use it for automotive at all. I wanted to use it to read out 2 RGB-photodiodes + 2 normal photodiodes synchronously to check lighting assemblies, their PWM drivers and DMD systems. I already tried to work with Audio-ADCs but didn't manage to run more than one of them in a synchronous and stable way.

For professional usage I borrowed a picoscope that worked like a charm. Probably that´s the best way to go.

PS: This is the result of all 8 channels in parallel. Now the mis-compensation is not that visible anymore because the horizontal resolution is crap. You don't even get 10kHz of bandwidth out of this thing...

[bobxyz]

Do people in automotive workshops really only want the timing information and signal shape just doesn't matter?

For simple automotive diagnostics, all you really care about is that the sensor / component has a somewhat reasonable waveform.  You're looking for shorted wires and failed components, not analyzing electrical margins or performance on a new design.  For automotive techs that are moving up from just using a multimeter, a simple scope with very limited bandwidth is a plus.  If the scope has 1MHz+ bandwidth, it will pick up noise and be confusing.  With a low bandwidth scope, the ground can be connected almost anywhere.

For intermittent automotive problems, stepping up to a real scope with deep storage is needed.  PicoScope is great, but the models with their automotive scope software are expensive.

In my closet is an old LeCroy 9362c with a couple active probes, but there's no way I'd haul it out to the garage and get anywhere near those 30KV+ automotive signals.

[I also have a long writeup elsewhere on using the sub $40 Fnirsi DSO152 pocket scope for auto diags - it's a fantastic choice as an introductory scope for backyard mechanics.]