Turn your Rigol DS1052E Oscilloscope into a 100MHz DS1102E
in [Design]
|
From: John Tserkezis on 31 Mar 2010 19:48 John Larkin wrote: > Since the ADCs are overclocked, That's only done to cut costs. If component selection *was* an issue, it would have been locked in hardware. A jumper perhaps, a bridge on a track, something. > it may be that Rigol selects the best scopes to be the 100 MHz versions. I don't think so. The testing would have occurred during the board construction phase, before it has been programmed with firmware and fit into a case. It appears the ONLY difference between the 50/100Mhz version is one character in the serial number, via unencrypted, keyboard-capable RS232 communications, on a port that's user accessible. Oh, and the sticky label. Call it what you want, it's sloppy, they *have* been caught with their pants down.
From: John Larkin on 31 Mar 2010 19:49 On Thu, 01 Apr 2010 08:17:52 +1000, Tom <tom(a)no.spam.invalid> wrote: >John Larkin wrote: > >> >> Since the ADCs are overclocked, it may be that Rigol selects the best >> scopes to be the 100 MHz versions. > >John, > >I cannot understand your logic - it is ok for Rigol to overclock slow >ADCs and deprive ADC makers from income but it is not ok to "overclock" >a Rigol scope? > >Tom There's nothing illegal about overclocking, or exploding, any chip. It might void a warranty, as power dissipation probably rises with clock rate. It may be illegal to use a computer to hack firmware if it deprives the IP owner of revenue. I just tested some plain vanilla 10 ohm 1206 resistors for overload capability. They die at 26 volts in 1 millisecond, 1 Hz rep rate. That's 2.6 amps, 68 watts. They are cool to watch on the IR viewer, sort of like a pulsing LED. So it looks safe to use them at 22 volts for 120 microseconds, which is what I need. I don't think anybody will sue me for resistor abuse. John
From: Jamie on 31 Mar 2010 20:55 John Fields wrote: > On Wed, 31 Mar 2010 11:19:08 -0700, John Larkin > <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: > > >>On Wed, 31 Mar 2010 13:08:45 -0500, John Fields >><jfields(a)austininstruments.com> wrote: >> >> >>>On Wed, 31 Mar 2010 09:30:03 -0700, John Larkin >>><jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >>> >>> >>>>On Wed, 31 Mar 2010 12:19:00 -0400, Spehro Pefhany >>>><speffSNIP(a)interlogDOTyou.knowwhat> wrote: >>>> >>>> >>>>>On Wed, 31 Mar 2010 08:53:03 -0700, John Larkin >>>>><jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >>>>> >>>>> >>>>> >>>>>>Why Jones would choose to hurt Rigel is a mystery to me. >>>>>> >>>>>>John >>>>> >>>>>What makes you think he hurt Rigol? They've have probably just sold >>>>>dozens of scope to people who wouldn't have otherwise bought a scope >>>> >>>>>from a Chinese maker. >>>> >>>>>Most companies will continue to buy what's guaranteed. >>>>> >>>>>He might have hurt or helped them. >>>> >>>>I'm sure that some people who would have bought the 100M version will >>>>buy the 50 and hack it. Not many, I expect, mostly amateurs. >>> >>>--- >>>So now it's _not_ "serious money" like you originally claimed? >> >>If it's, say, 100 scopes hacked at a loss of $400 each, until Rigol >>makes the firmware more secure (which will also cost money to do) >>that's $40K. I don't know if $40K is "serious" money that matters to >>Rigol, or to you. $40K is fairly serious to me. > > > --- > Jeez, John, I see you still haven't quit being a cheater... > > 1052E's go for $595 max _retail_, and 1102E's go for $795 max, also > retail, so that's a difference of $200, of which Rigol sees maybe $50. > > Applied to 100 scopes, that's $5K which is probably chump change for the > likes of Rigol. > > Now if I cheated a little and claimed that those 100 scopes would never > have been bought except to be "converted", then I could claim that the > extra sales more than offset any losses (especially since it costs them > the same to build either scope) and that the hack was actually a > blessing in disguise, if not leaked on purpose... > --- > > >>How would you feel if Jones hacked one of your products and cost you >>$40K? But I think you don't do firmware, so the question is probably > > > --- > If he hacked one of my products and wasn't in violation of any IP > restrictions, then I'd be unhappy but that's the way it goes... > > BTW, you think wrong. > > Again. > > I do hardware, firmware, software, AND bleeding edge 555 circuit design. > > So there... > > JF :)
From: John Larkin on 31 Mar 2010 19:55 On Thu, 1 Apr 2010 00:26:58 +0200, "fritz" <yaputya(a)microsoft.com> wrote: > >"John Larkin" <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in message >news:oph7r51vibegk37bkncrn8avtiou3p6ssk(a)4ax.com... >> On Wed, 31 Mar 2010 23:30:18 +0200, "fritz" <yaputya(a)microsoft.com> >> wrote: >> >>> >>>"John Larkin" <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in >>>message >>>news:mtq6r5t2e14htcdl9svbr3bt8g95hlpmmc(a)4ax.com... >>>..... >>>> Looking at the transient response at 100 MHz, which kinda sucks, I >>>> wonder if the 50 and 100 MHz scopes are indeed identical except for >>>> firmware. >>>> >>>> John >>> >>>Kinda sucks ? >>>Did you watch the eevblog ??? I don't think you have the slightest clue >>>about >>>what fast signals really look like. >> >> >> How about this one: >> >> ://www.highlandtechnology.com/DhttpSS/T760DS.html >> >> That's a real transformer-isolated 100 volt pulse into 50 ohms. We've >> tweaked it since we took that pic, and rise/fall are now typically >> under 1 ns. >> >> And this is a 1 GHz square wave >> >> http://www.highlandtechnology.com/DSS/T860DS.html >> >> The undershoot is my fault... a trace is a little too long. I'll fix >> it next pass. >> >> The higher the bandwidth the messier >>>they look as various resonance effects in the measurement circuit >>>are revealed - use a 1Ghz 'scope and they REALLY suck. >> >> I use a 20 GHz scope, and the calibration and TDR pulses are almost >> perfect. >> >> John > >You claimed the modded Rigol 'kinda sucks'. The step response is pretty ugly. The step response of my unmodded 1052E is very nice. >Why ? What were you expecting from a 100Mhz scope ? A nice Gaussian step? >You also snipped the following... >"The modded Rigol compared very well with a 100Mhz Tektronix TDS 1012." >Care to comment why a TDS 1012 also 'kinda sucks' ? > > Compared to a clean scope, it does. They are made in China too. I don't know if Tek designs them or just rebrands. I also don't know how clean a signal Dave applied. John
From: John Tserkezis on 31 Mar 2010 20:03
George Jefferson wrote: > Your justifications only show that you fit in the same group as Rigol. I > won't be buying anything from you and I hope your customers will find > someone else to give their money to. Perhaps he doesn't appreciate that Rigol make these things in *quantity*. Remember, they're overclocking the ADCs, so they're cutting costs in every way they can. His justification would only work on small runs, where it would worthwhile to have someone MANUALLY plug in a port, test, program accordingly, and stick the relevant label. The only cheap way to do it, is to have a production line test jig, that automagically programs, probes and accordingly presets the equipment to the required spec. All of this would be most economically viable for the numbers Rigol are handling, and even EASIER to obscure a simple "one character" difference. Instead, they have this part of the communications available to the end user, via an industry standard interface and keyboard-capable commands. Mind you, they *could* be doing it as I outlined above. But the fact remains, if it were so easy to obscure this part from the end user, why did they almost appear to go out of their way to make it easy instead? Whether or not the user re-programming is, or is not allowed, likewise if it should or should not be allowable, is rather irrelevant now. Whatever the reason, they *HAVE* made it easy, and this pretty much negates any whining IMO. That bolted horse comes to mind. |