A while ago my DS4014 started acting up. Rail-to-rail noise on channels 3 and 4 which could be mitigated by waiting an ever-increasing amount of time after cold startup or by a magic finger poke to ADC 3
. Eventually the wait became infinitely long, pushing me to attempt a fix. I succeeded by replacing ADCs 3 and 4 ("Rigol" brand) with two ADC08D1000 ordered from taobao (I was hoping to get MXT2001s but it looks like I got the real deal instead). This *almost* worked but had artifacts which could only be fixed by DC coupling the CLK lines to the replaced ADCs.
Full story + pictures:
https://goo.gl/photos/dxU3ChWUcvCMDW4N9Why DC couple the clocks? Because it looks like the "Rigol" ADCs differ from both ADC08D1000 and MXT2001 in that they lack 100 ohm CLK termination -- and Rigol takes advantage of this fact by suspending the AC coupled CLK while synchronizing ADCs 1,2 and 3,4. With no termination, this works fine. With 100 ohm termination, it (dis)charges the AC coupling capacitors until CLK+ and CLK- droop into each other and trigger random clocking, upsetting the phase relationship the scope was trying to set up in the first place.
Anyway, I hope there aren't many of you with the "dying DS4014 that Rigol wants $930+shipping to repair" problem, but if there are, I hope this helps
EDIT: It almost goes without saying, but be wary of out-of-warranty DS4000s (or 90-day warranty from the official bargain bin). I've heard other people mention a similar problem to the one I had, and we know that Rigol recently changed the DS4000 line to add internal heatsinks on the ADCs. Their absence might have caused my problem and it might cause you a problem if you get a DS4000 that has lacked heatsinks for all or most of its life.
Wow, nice job. You have some nice kit, too.
And I love the upside-down gravity-assist soldering technique. Very nice!
Great job with the rescue, there.
Wow, nice job. You have some nice kit, too.
And I love the upside-down gravity-assist soldering technique. Very nice!
Great job with the rescue, there.
Thanks!
Yeah, we've got some sweet kit at work, no doubt about it. I don't normally get to play with it since I'm a software guy so this was a fun experience, especially the Tek 5204. 400Gs/s (ET, but still)! Those probes, too... I never knew how much of a joy sharp springy tips could be. Having the right kit makes all the difference.
Fancy debugging and a good writeup! That heatmap alien snot really is a fine looking mess of a signal
EDIT: It almost goes without saying, but be wary of out-of-warranty DS4000s (or 90-day warranty from the official bargain bin). I've heard other people mention a similar problem to the one I had, and we know that Rigol recently changed the DS4000 line to add internal heatsinks on the ADCs. Their absence might have caused my problem and it might cause you a problem if you get a DS4000 that has lacked heatsinks for all or most of its life.
Cool repair story. My DS4014 from their bargain bin came with only one heatsink on the main FPGA. Everything else was open in the air (the picture in
reply 393 of the other DS4000 thread).
Very well done!
Interesting reading and nice pics, thanks.
Thanks for the kind words, everyone.
Cool repair story. My DS4014 from their bargain bin came with only one heatsink on the main FPGA. Everything else was open in the air (the picture in reply 393 of the other DS4000 thread).
Yikes! I hope it doesn't cause you any BGA rework going forward.