0 Members and 1 Guest are viewing this topic.
Looks like Keysight engineers understand PLL design much better than Rigol engineers! https://www.eevblog.com/forum/testgear/agilent-dsox2024-won't-boot/https://www.eevblog.com/forum/testgear/oh-no!-i'm-never-buying-any-agilent-keysight-equiment-ever-again!/
Quote from: Fungus on May 22, 2016, 04:29:35 pmLooks like Keysight engineers understand PLL design much better than Rigol engineers! https://www.eevblog.com/forum/testgear/agilent-dsox2024-won't-boot/https://www.eevblog.com/forum/testgear/oh-no!-i'm-never-buying-any-agilent-keysight-equiment-ever-again!/Of course they do. They detect if there is no PLL lock. Rigol just silently outputs the variable frequency jittery crap into the ADC clock.
Quote from: wraper on May 25, 2016, 07:06:04 amQuote from: Fungus on May 22, 2016, 04:29:35 pmLooks like Keysight engineers understand PLL design much better than Rigol engineers! https://www.eevblog.com/forum/testgear/agilent-dsox2024-won't-boot/https://www.eevblog.com/forum/testgear/oh-no!-i'm-never-buying-any-agilent-keysight-equiment-ever-again!/Of course they do. They detect if there is no PLL lock. Rigol just silently outputs the variable frequency jittery crap into the ADC clock.But it seems to still meet its specs? Are you saying it was over-engineered?
Quote from: hendorog on May 25, 2016, 07:20:19 amQuote from: wraper on May 25, 2016, 07:06:04 amQuote from: Fungus on May 22, 2016, 04:29:35 pmLooks like Keysight engineers understand PLL design much better than Rigol engineers! https://www.eevblog.com/forum/testgear/agilent-dsox2024-won't-boot/https://www.eevblog.com/forum/testgear/oh-no!-i'm-never-buying-any-agilent-keysight-equiment-ever-again!/Of course they do. They detect if there is no PLL lock. Rigol just silently outputs the variable frequency jittery crap into the ADC clock.But it seems to still meet its specs? Are you saying it was over-engineered? Who meet the specs, Rigol? Certainly was not the case For DS1054Z because it had apparent effects on the waveform captured. Until they firmware(ish) kinda "almost" fixed non locking PLL, almost because the clock still remained not that clean, and it's not certain if 100% of the scopes became good enough to not have visible issues. This varies from scope to scope and you don't know how bad your particular unit is.As of Keysight, scope checks and informs you if PLL gone wrong, therefore you know when the scope cannot be trusted.
I think the DS1k problem was only visible once the data was extracted from the scope and an FFT run outside the scope? Perhaps that is also the case here.
Quote from: hendorog on May 25, 2016, 07:51:31 amI think the DS1k problem was only visible once the data was extracted from the scope and an FFT run outside the scope? Perhaps that is also the case here.No, it was perfectly visible on the scope screen. Dave even made a video about this.
It was discussed on page 4 of this thread - about DS2k not the earlier issue about DS1k.
Quote from: hendorog on May 25, 2016, 07:51:31 amIt was discussed on page 4 of this thread - about DS2k not the earlier issue about DS1k. You must have missed following this link in the OP:https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg552777/#msg552777
Quote from: tautech on May 25, 2016, 08:08:28 amQuote from: hendorog on May 25, 2016, 07:51:31 amIt was discussed on page 4 of this thread - about DS2k not the earlier issue about DS1k. You must have missed following this link in the OP:https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg552777/#msg552777Yep but I believe also prior to the update which resolved it to the point where it behaved normally.Buds investigation was on what was left behind after the update.
Quote from: hendorog on May 25, 2016, 08:15:17 amQuote from: tautech on May 25, 2016, 08:08:28 amQuote from: hendorog on May 25, 2016, 07:51:31 amIt was discussed on page 4 of this thread - about DS2k not the earlier issue about DS1k. You must have missed following this link in the OP:https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg552777/#msg552777Yep but I believe also prior to the update which resolved it to the point where it behaved normally.Buds investigation was on what was left behind after the update.I ask you how can a FW update fix a PLL design error, it can't, it only masks the problems it creates.Buds investigation follows on from what was discovered in the DS1054Z WITH a DS2072A and confirms there's been an error in design, so bad that the PLL manufacturers datasheets layout recommendations appear to have been totally ignored.Look, all manufactures make mistakes and there's many examples in threads here, this is just another one.
Quote from: tautech on May 25, 2016, 08:26:08 amQuote from: hendorog on May 25, 2016, 08:15:17 amQuote from: tautech on May 25, 2016, 08:08:28 amQuote from: hendorog on May 25, 2016, 07:51:31 amIt was discussed on page 4 of this thread - about DS2k not the earlier issue about DS1k. You must have missed following this link in the OP:https://www.eevblog.com/forum/blog/eevblog-683-rigol-ds1000z-ds2000-oscilloscope-jitter-problems/msg552777/#msg552777Yep but I believe also prior to the update which resolved it to the point where it behaved normally.Buds investigation was on what was left behind after the update.I ask you how can a FW update fix a PLL design error, it can't, it only masks the problems it creates.Buds investigation follows on from what was discovered in the DS1054Z WITH a DS2072A and confirms there's been an error in design, so bad that the PLL manufacturers datasheets layout recommendations appear to have been totally ignored.Look, all manufactures make mistakes and there's many examples in threads here, this is just another one.Yep a firmware upgrade _can_ resolve a PLL error. I know this because I just built one and the relationship between the physical components and the settings was important enough to make the difference between not locking, locking with massive spurs, and locking with minimal spurs. (that was the best I could do on that board because I ignored the manufacturers advice, new PCB on the way
I completely agree with your second point - they screwed up. But Keysight screwed up too, the only difference was that they detected their screwup and charged $2k to fix it...
But Keysight screwed up too, the only difference was that they detected their screwup and charged $2k to fix it...
detected their screwup and charged $2k to fix it...
Quote from: hendorog on May 25, 2016, 08:35:51 amQuote from: tautech on May 25, 2016, 08:26:08 amI ask you how can a FW update fix a PLL design error, it can't, it only masks the problems it creates.Buds investigation follows on from what was discovered in the DS1054Z WITH a DS2072A and confirms there's been an error in design, so bad that the PLL manufacturers datasheets layout recommendations appear to have been totally ignored.Look, all manufactures make mistakes and there's many examples in threads here, this is just another one.Yep a firmware upgrade _can_ resolve a PLL error. I know this because I just built one and the relationship between the physical components and the settings was important enough to make the difference between not locking, locking with massive spurs, and locking with minimal spurs. (that was the best I could do on that board because I ignored the manufacturers advice, new PCB on the way Well we learn something every day....thanks.Do consider sharing your Yaigol design *moment* in another thread, I'm sure we'll all learn from your experience.
Quote from: tautech on May 25, 2016, 08:26:08 amI ask you how can a FW update fix a PLL design error, it can't, it only masks the problems it creates.Buds investigation follows on from what was discovered in the DS1054Z WITH a DS2072A and confirms there's been an error in design, so bad that the PLL manufacturers datasheets layout recommendations appear to have been totally ignored.Look, all manufactures make mistakes and there's many examples in threads here, this is just another one.Yep a firmware upgrade _can_ resolve a PLL error. I know this because I just built one and the relationship between the physical components and the settings was important enough to make the difference between not locking, locking with massive spurs, and locking with minimal spurs. (that was the best I could do on that board because I ignored the manufacturers advice, new PCB on the way
I ask you how can a FW update fix a PLL design error, it can't, it only masks the problems it creates.Buds investigation follows on from what was discovered in the DS1054Z WITH a DS2072A and confirms there's been an error in design, so bad that the PLL manufacturers datasheets layout recommendations appear to have been totally ignored.Look, all manufactures make mistakes and there's many examples in threads here, this is just another one.
QuoteI completely agree with your second point - they screwed up. But Keysight screwed up too, the only difference was that they detected their screwup and charged $2k to fix it...But that's not over yet.....once the word get out more and other user/owners that have had to pay for the fix find out....shit anything could happen, court cases, recalls, hey it's all happened to manufacturers before.
Quote from: hendorog on May 25, 2016, 08:35:51 amBut Keysight screwed up too, the only difference was that they detected their screwup and charged $2k to fix it...There is a big difference between screwed up design and faulty particular unit.
Yes, but that is on a 2.8k scope. Which is not commensurate with an A brand supplier. If you have to throw away your scope and buy a new one when it breaks then there is no point buying A brand gear.
2- Taxes which focuses and how to calculate?Import Tax: 60% of the Customs Valuation (Customs Value = the value of goods sum + shipping)ICMS: on average 18%, varying according to the State
There is a big difference between screwed up design and faulty particular unit. Also, don't forget about that it is Brazil, prices will be 2x+ of the US.Quotedetected their screwup and charged $2k to fix it...After warranty has already expired.
Quote from: wraper on May 25, 2016, 08:59:13 amThere is a big difference between screwed up design and faulty particular unit. Also, don't forget about that it is Brazil, prices will be 2x+ of the US.Quotedetected their screwup and charged $2k to fix it...After warranty has already expired.Warranty aside, this really looks like some sort of serious design defect. Keysight could be in big trouble if more people read this thread and say, "That happened to me, too!".
Quote from: Fungus on May 25, 2016, 12:21:40 pmWarranty aside, this really looks like some sort of serious design defect. Keysight could be in big trouble if more people read this thread and say, "That happened to me, too!".Seriously? PLL fault was not a design defect. All it was unlucky to receive such a scope and not test it in reasonable time period. And WTF Keysight out of warranty discussion is doing in the Rigol tread.
Warranty aside, this really looks like some sort of serious design defect. Keysight could be in big trouble if more people read this thread and say, "That happened to me, too!".
Quote from: wraper on May 25, 2016, 12:50:43 pmQuote from: Fungus on May 25, 2016, 12:21:40 pmWarranty aside, this really looks like some sort of serious design defect. Keysight could be in big trouble if more people read this thread and say, "That happened to me, too!".Seriously? PLL fault was not a design defect. All it was unlucky to receive such a scope and not test it in reasonable time period. And WTF Keysight out of warranty discussion is doing in the Rigol tread.The 'scope was tested, placed in storage for a while, then refused to boot.If the firmware is somehow being corrupted (for whatever reason, including "putting it in storage for three months") then that's a design defect.Keysight has already admitted they've changed the firmware in more recent models so it's easier to re-flash it if it fails. That sounds to me like they've seen a problem.
Maybe you should read it one more time?
Quote from: wraper on May 25, 2016, 01:12:48 pmMaybe you should read it one more time?Maybe I should, it's hard to remember which of all the failed Keysights 'we're actually talking about.