Rigol 2000 series (ds2072) Teardown and repair

[playfsx]

Well here is the video :

[EEVblog]

Damn, that was going to be my Teardown Tuesday tomorrow!

Dave.

[A Hellene]

Do you realise that announcing the release of such an interesting presentation before it actually becomes accessible, makes you both cruel sadists?


-George

[SeanB]

You got done over Greek style........................

[zaoka]

Damn, that was going to be my Teardown Tuesday tomorrow!

Dave.

I would love to see features exploring more than teardown...

[Astroplio]

Hey what happen there?
The heatsink came loose because the hook came off???
The little U clip gave in? Too much heat caused it to desolder or really bad soldering in the first place, what do you think?

Oh man I got the same model just last week, now you got me worried...

[nctnico]

An anti-static rubber mat would help to avoid scratches and ESD problems.

[playfsx]

An anti-static rubber mat would help to avoid scratches and ESD problems.

Yep that it's correct but i don't have one yet , im thinking to buy one if you have something to suggest ..
I found this one http://www.ebay.co.uk/itm/Professional-Antistatic-Mat-ESD-PC-repair-earth-clip-/400281021082?pt=LH_DefaultDomain_3&hash=item5d329baa9a#ht_4177wt_1163
But i don't know about the quality .

Hey what happen there?
The heatsink came loose because the hook came off???
The little U clip gave in? Too much heat caused it to desolder or really bad soldering in the first place, what do you think?

Oh man I got the same model just last week, now you got me worried...

I have to say 3 things about the teardown
First take apart this scope is relay easy all the screws its the same and in every hole need one the have a indication.
The motherboard seems really nicely design and the soldering seems like a first class.
Except the hand soldering parts this is where the blow it up , i found cold joints and because of that probably the clip came off .
I don't thing that the heat its melt the solder, when i try to resolder it i put the soldering station in 480C to melt it.
In the video seems that its only one clip drop from the scope but accusal it's two clips that are missing , both the heat sings it's stuck in the chips with only the thermal paste and with
any vibration it can came off damaging the scope , talking about damage i lucky that the floating metallic parts don't make any damage in the first place.
Thirdly, in the scope i found 1500mbit of ddr2 ram , i mean im sure that this scope it's capable to make more than 54mpt capture,  in the record mode you can record 2 waveforms of 54mpt so this is 108mpts hmmmm ..... maybe with some sort of hack we can unlock the true power of this scope

Astroplio probably the your scope it will be fine just try to shake it and hear for any noise if it's make just sent it back.
But let me tell you two more thing, fist careful with the probes it's relay not that good, it's the second time that my probe is cut and im not the kind of user that i don't care my staff and second DO NOT make a auto calibration, it will expire the trial version modules.


What you guys thing about the power supply it's any good ?? 
 
edit: 1500mb -> 1500mbit

[EEVblog]

Thirdly, in the scope i found 1500mb of ddr2 ram , i mean im sure that this scope it's capable to make more than 54mpt capture,  in the record mode you can record 2 waveforms of 54mpt so this is 108mpts hmmmm ..... maybe with some sort of hack we can unlock the true power of this scope

My teardown is uploading now (if it work, youtube issues...).
Yes, I found 2 x 64MB DDR2 SDRAM chips (one for each channel presumably) on the acquisition FPGA. A third chip of the same size and type is on the display processor FPGA. So yes, 1.5Gbit total (192MB)

What you guys thing about the power supply it's any good ??

It looks excellent quality, best you could expect.
It's branded Rigol and no third party markings.

Dave.

[playfsx]

My teardown is uploading now (if it work, youtube issues...).

Im waiting for it

Yes, I found 2 x 64MB DDR2 SDRAM chips (one for each channel presumably) on the acquisition FPGA. A third chip of the same size and type is on the display processor FPGA. So yes, 1.5Gbit total (192MB)

Oh i write mb instead of mbit i will fix my post.
So this is 8bit resolution scope it be normally take 1byte per sample then for 54mpts it will be need 54mbyte of memory so they use the half memory , i really don't like that limit the capability of this machine in software , of courses if they give in the 2000 series 100+ mpts it will competes with the 4000 series....well...
 
Dave if i short a little the cable of the probe (for fixing it) do you thing it's going to effect the signal ?

[GEuser]

I like the Speed Screwdriver , what brand is it 

Personally i'd feel a bit shortchanged , for 945us and no Finger marks all over the board , WHAT!

[saturation]

While the teardown was good, it lacks a learned commentary particularly the areas that look like potential faults.  The build seems high overall, but there are some hand solder portions that need clarification to the public, and I'm not clear on what those clips that fell off were, where were they?  Did they hold the heatsinks on the case when the 2 halves are reassembled?  Did it occur only on disassembly or was it already loose and shaking inside?  That kind of stuff.

Damn, that was going to be my Teardown Tuesday tomorrow!

Dave.

[firewalker]

Have you notified Rigol for the issue?

Alexander.


Apo pou to paragiles?

[A Hellene]

Dave if i short a little the cable of the probe (for fixing it) do you thing it's going to effect the signal ?

Do not touch the oscilloscope probes! You will either destroy them or alter their characteristics!

[...]
The oscilloscope probes are extremely complicated animals, despite the textbooks that describe them as simple passive devices with resistive and capacitive dividers. For example, no x1 passive probe can go further beyond the 20..50 MHz barrier. Why?
Please, read an excellent article by Doug Ford on the oscilloscope probes, that was published at the Silicon Chip in 2009:
The secret world of Oscilloscope Probes.
[...]

(Corrected the invalid original article link)

You can also read THE classic oscilloscope probe manual by Tek, Oscilloscope Probe Circuits (Joe Weber, 1969).


-George

[playfsx]

Have you notified Rigol for the issue?

Alexander.


Apo pou to paragiles?

Nop i don't send a e-mail to rigol , i order from German store called Batronix

Dave if i short a little the cable of the probe (for fixing it) do you thing it's going to effect the signal ?

Do not touch the oscilloscope probes! You will either destroy them or alter their characteristics!

The probe from ch1 is all ready damage so i don't have nothing to lose try to repairing it, so i remove 14cm of the cable i solder it and the probe work again, probably i alter the characteristics a little but i can't detect any different with the 1khz built in signal either in 1x or 10x mode, compare to the other probe in the same channel.
Because i don't have they equipment to generate signal more than 500khz i can't do more scientific test .
When i get i nice function generator i will test it more , at the moment i use the other probe for critical staff.

[firewalker]

Sending it back for service (warranty) or as a faulty unit, wasn't an option?

Alexander.

[playfsx]

Sending it back for service (warranty) or as a faulty unit, wasn't an option?

Alexander.

If a know that the u clips it came off before i open it, i will sent it  back, now i open it and fix it there is not reason to do so,

btw hellene i test the probes with 312,8khz square wave and i find that the repair probe have a slight more fall time(26,7ns) than the other(23,7ns), ofcourses i alter the channels to confirm that.
Is 3ns different a big difference so i must buy a new probe ?

[A Hellene]

Since an oscilloscope probe is not just a piece of wire, but a balanced transmission line, the question now is, how much can you trust an out-of-balance transmission line that a shortened/modified probe has now become?

Please, read the documents I have linked to in my previous message.


-George