A competitor for the Rigol?

[bogdan546]

I was curious what it looks like

[cybergibbons]

Can you check out the SD card slot?  Any issues with various capacities below and above 2GB?
Does it support FAT16 and FAT32 formatted cards?

The Rigol DS1052E seems to work with pretty much any FAT formatted USB stick.
The Hantek DSO-1060 I have doesn't like FAT format, only FAT32, not a big deal
but an undocumented annoyance.

Scott

I've not tried FAT16, but the GDS-1152A works fine with 4Gb cards. Painfully slow to transfer whole memory to it though.

[McPete]

Can you check out the SD card slot?  Any issues with various capacities below and above 2GB?
Does it support FAT16 and FAT32 formatted cards?
Scott

I just tried an 8Gb SD HC, formatted FAT32- No problems at all. Scope recognises it quickly, stores a snapshot. The included SD card reader also works with my OS 10.5 Macbook straight up, no issues.

 I was a little worried that it wouldn't support SD HC. So far as I can see, it just works! Given that a snapshot is 32Kb or so, anything more than a 1Gb card is most likely overkill!

please make a movie with rectangular signal to calibrate the probes
 to me it is so. how is you?

Bogdan,

The top of the wave should be flat and "square"- Like the wave on the right of your second picture. Anything else means you still need to adjust your probes. I can take a movie if you REALLY want me to, but it isn't really going to show you anything new.

P.

[bogdan546]

so the vertical line is not one continuous piece

[FreeThinker]

so the vertical line is not one continuous piece

This is more likley due to the resolution of the display.Get a piece of graph paper and draw a line on it at an arbitrator angle ie not Horizontal or vertical.Look closly and you will see that the line sometimes is contained within the square and sometimes it covers two squares,your screen cannot light up half a pixel so has to decide which one is closer to the ideal line hence the offset or 'Jaggies' .Try zooming in and you may notice a improvement.

[bogdan546]

This is more likley due to the resolution of the display.Get a piece of graph paper and draw a line on it at an arbitrator angle ie not Horizontal or vertical.Look closly and you will see that the line sometimes is contained within the square and sometimes it covers two squares,your screen cannot light up half a pixel so has to decide which one is closer to the ideal line hence the offset or 'Jaggies' .Try zooming in and you may notice a improvement.
[/quote]
Thank you for explanations

[saturation]

I'm happy for you.  If you could tell us how well it works:

Frequency response to its rated 60Mhz

How well its fast fourier mode looks like before the noise floor makes reading the harmonics unusable

ENJOY! and thanks for posting.

If there's anything in particular anyone would like me to try with it, please let me know and I'll endeavour to get back to you with a result.

P.

[McPete]

If you could tell us how well it works:

Frequency response to its rated 60Mhz

How well its fast fourier mode looks like before the noise floor makes reading the harmonics unusable

ENJOY! and thanks for posting.

1 & 2) This will happen on Tuesday, barring my forgetfulness.


3) No worries mate, glad to be of service.

[saturation]

Thanks a gigabunch McPete, those are the major features it differs from a stock 1052E Rigol.

If you could tell us how well it works:

Frequency response to its rated 60Mhz

How well its fast fourier mode looks like before the noise floor makes reading the harmonics unusable

ENJOY! and thanks for posting.

1 & 2) This will happen on Tuesday, barring my forgetfulness.


3) No worries mate, glad to be of service.

[McPete]

Sorry for the delay folks, here, slightly belatedly, are some snapshots from my testing of my 1062A. The standard used, if you're interested, is a Fluke 5520A with the scope option, although a few may be with function generators.

First image; 400mV P-P sine wave at 20MHz, with FFT as well. Note, the fundamental and first three harmonics are pretty clear. Note; This signal was produced using a function generator.

Second image; using cursors on the same signal as above, to measure the FFT reading; X2 (the solid red cursor) is on the 5th harmonic, which is still discernible above the noise. Note; This signal was produced using a function generator.

Third image; 60Mhz Sinewave, 30mV P-P no FFT.

Fourth image; 100Mhz sinewave, 30mV P-P no FFT.

Fifth image; 200MHz sinewave, 30mV P-P, no FFT.

Sixth image, 30Mhz sinewave, 30mV P-P, no FFT.


If anyone would like more specific tests done, please let me know and I'll see what I can do.

McPete

[saturation]

Thanks a Gogol for the quick tests, McPete, its reads like the noise floor is just as noisy as the Rigol.

Could you run the same tests you did but use a square wave instead?  It would qualitatively show the bandwidth and any nonlinearity.  The sine wave tests show it rolls off gently and seems useful for measurements up to 100MHz.

At least look at 6 MHz [ which should be quite rectangular] versus 10 MHz,  the roll off on the edges would be pretty obvious [ as the edges represent the 9th harmonics or 54 Mhz and 90 MHz respectively.]

[DJPhil]

A big thanks to McPete, and to all those who contributed for all the hard work.
For me at least, the Instek is really looking good. I think I'll be ordering one when I finish saving up.

[colinbeeforth]

saturation said:

"How well its fast fourier mode looks like before the noise floor makes reading the harmonics unusable"

Yes, most FFTs on most DSOs are poor like this.  The problem is this - 8 bits vertical resolution = dynamic range of only 20log(10)255 or 48dB.  Since most fast acquisitions only have effective bits around 6 bits (not 8 bits, caused by noise and timing errors), then the effective dynamic range is only about 35dB in reality.  No Chinese manufacturer quotes effective bits at speed, so new buyers have no idea what it is.  In practice FFT is only of very limited use in modern DSOs.  Earlier LeCroy DSOs had a nice averaging mode which allowed you to trade off horizontal samples for vertical samples.  It only worked for lower frequencies where you could be sure of horizontal over sampling, and that is only possible if you have long acquisition memory, you could average the signal and achieve up to 3 bits of vertical enhancement.  It was based on sub sampling by counting the number of times natural noise flipped a sampled point into upper and lower bins.  Complex maths, but the end result for lower frequency signals was the equivalent of 11 bits vertical, so FFT could be more useful.  To get this mode, you needed: slow speed signals much lower than max sample rate, long memory, advanced math processing, the ability to chain math processing ending in FFT.  It made the scope run a bit slow but gave useful results.  Modern cheap DSOs simply can't do that stuff, the acquisition memory is too short to start with.  Weirdly, the LeCroy had a Motorola 68020 processor, so modern DSOs should be able to beat that, but it doesn't seem that Chinese scopes have got to the point of doing any real clever signal processing.  Maybe there just isn't enough demand.

[saturation]

Thanks for the insights, Colin.  I see your posts on the Hantek/Tekway thread which also has similar limits to its FFT, given the theoretical limits of 8 bits and hardware, do you have the facilities to actually measure the real noise floor of your Tekway?

saturation said:

"How well its fast fourier mode looks like before the noise floor makes reading the harmonics unusable"

Yes, most FFTs on most DSOs are poor like this. ... Weirdly, the LeCroy had a Motorola 68020 processor, so modern DSOs should be able to beat that, but it doesn't seem that Chinese scopes have got to the point of doing any real clever signal processing.  Maybe there just isn't enough demand.

[pablo]

hello McPete

Would you mind to tell how much ripple noise (peak to peak) your oscilloscope has when you short out the probe to ground?
I have one of those Instek GDS1062A oscilloscopes and mine shows a bigger ripple when probe's switch is on 10x.

I would really appreciate it if you test both positions (1x and 10x)

thank

[alm]

Would you mind to tell how much ripple noise (peak to peak) your oscilloscope has when you short out the probe to ground?
I have one of those Instek GDS1062A oscilloscopes and mine shows a bigger ripple when probe's switch is on 10x.

Sounds like inductive pickup in the loop between the tip and the ground. The voltage would increase for equal current if you increase the input impedance. Try minimizing the loop area by shorting the probe tip to the ground collar with something conductive like aluminium foil.