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.