Looks like Instek has some new models out now with 1GSa/s to replace the GDS-800 series. The GDS-2000(all color) series has a USB port in the front for flash drive and one in the back for printing. Only a 3 year warranty though, but the price is better than the Tek. It also has a TFT screen which should look better in bright light if I'm not mistaken. The GDS-2102 is US$1165 ($1295 msrp) for 2-channel/color/100MHz/RS-232/USB
Now I'm checking out the Rigol stuff. Looks like the Rigol 1000 series is about 5-6" deep like the Tek, but has only 400MSa/second, but does have 1M (yes that's an M) points. They also claim a TFT display. The Rigol 5000 series is about a foot deep (like the Instek) but has 1GSa/second ability. They don't specify the display type, so I assume that means that it's not TFT. Looks to have the typical set of features, maybe a bit richer set than the Tek? The DS1102C (100MHz/2-channel) is catching my fancy right now at $999. The logic-analyzer (MSO?) version (DS1102CD) would really be nice, but for the money, I think the Intronix LogicPort might be a better way to go for that functionality.
Anyone else used a Rigol? Does anyone know if the 5000 series is newer than the 1000 series? The biggest downside I see to the Rigol right now is finding a distributor in the US. I only found these places:
formatting link
formatting link
Same pricing. Anyone dealt with either of them before?
Looks like it's coming down to the Instek GDS-2102 (100MHz/2-channel) or the Rigol DS1102C.
We have some of the current Agilent DC power supplies that are digitally controlled (wherein you set the regulated voltage/current using an encoder knob, you can memorize settings, there's a GPIB interface, etc.), and it makes several highly-visible birdies on a spectrum analyzer. :-( For RF boards I still use the older HP "all linear" power supplies... which I find nicer to use in the common case where you don't need to memorize 10 different settings.
One reason why this client of mine bought those "older" supplies on EBay as well. They are clean. Monday I almost did the usual, trudging over to the stationary room to get some C-cells I could solder in series when I glanced at the lab supply. Ah, it's an old analog one, I don't need to do the battery spiel here.
It's a lot easier to just snap a pic. Plus, I can add a post-it on the bezel to document the situation in the same pic. There's nothing worse than having 10 different pictures and not remembering what they represent.
The 7000 series are mighty fine scopes. Some day when I have time and find an old Polaroid enclosure I might build a digital camera setup that you could simply hinge onto the screen bezel. Somehow those kinds of pictures are more lively than the bland computer plots. Just like real picture in a real movie versus that animated stuff that the kids watch.
The Tek TPS2024 200 MHz digital scope is stunning. All four channels, and the external trigger, are individually electrically isolated. And battery power is standard, even more overkill. And it's a beautiful ergonomic design.
Imagine clipping your probe ground lead onto a bus that's 200 volts off ground, and looking at millivolt signals relative to that.
The 1000 mixed signal series is the newer series. The 5000 was their first offering. As mentioned in another post I have a Rigol 5102 in the lab, and whilst I haven't used it much, it is not a bad scope. It has some very nice features like digital filtering and masking, but you can't dump the screen to USB key, but I think you can on the newer 1000 series?
places:
formatting link
No contest, the 1M sample memory (512K/channel) of the Rigol wins hands down. The lower 400MS/s sample rate is still high enough (just).
The Instek beats the Tek with it's 25K memory, but both get crapped on by the Rigol. You can do sooooo much more with that extra memory, you won't ever regret it.
Lemme see, a $3000 4 channel Tek compared with a $2000 2 channel Agilent (that's really a $1000 Rigol in disguise), yep completely "fair and balanced" on that count. ;-) I didn't notice Tek pointing out how the Agilent can do almost twice the number of automatic calculations. ;-)
Examining the disclaimers, I'm of the opinion that some of their "facts" may have been changed in later Agilent/Rigol firmware releases. Granted the Tek has its advantages too. For example the Tek has an on-screen clock. That's gotta help John keep his post-it photographs in order. (;-) Seriously, I'm sure the Tek is a fine scope, it's just too rich for my blood.
Ok, after I did some more looking I figured it out. I didn't figure they would double the size of the case on a later model.
It is my understanding that yes, you can dump to a memory stick plugged into the front. There is a port in the back for a printer as well. What scope do you primarily use?
places:
formatting link
I kinda wish it had a higher sample rate, but most of what I do is so far removed from the upper bandwidth of the scope that I really don't think it should be an issue. Since I've never used a DSO before, I don't have much "feel" for all these numbers and their significance. I'm figuring it out though. :-) The comparisons on the TEK site are interesting and informative, even if they might be tilting things just a wee bit.
It might already be spent. ;-) Have you seen the Intronix LogicPort analyzer, it looks way cool with built in datastream decoding for SPI, I2C, serial ASCII and "more". 34 channels, should be more than enough for my needs.
The Rigol DS1102CD (note the 'D' on the end) has a 16 port logic analyzer. They want $500 for the option though. :-( The Intronix is less expensive and more functional (excepting that you could set up complex triggers with the analyzer built into the scope) Or do you think it would be wiser to go with the integrated analyzer.
My primary scope is an Agilent MSO6032A (mixed signal 300MHz), with an older model Agilent 54621D mixed signal scope as the secondary. The Rigol brings up the rear as a cheapie throw around scope. Had a Tek
3000 series as well, but it was lost in transit.
places:
formatting link
Ideally you want the sample rate to be at least 10 times the bandwidth, so you can get some half-decent waveform resolution at the bandwidth limit. So for a 100MHz scope you want 1GS/s.
400MS/s for a 100MHz scope is still ok, at least you won't be aliasing and you'll have a couple of points of detail at the full bandwidth. But I'd rate the Rigol as only having a 40Mhz single shot bandwidth in this case, compared with the 100MHz single shot bandwidth on a 1GS/s unit. But this a reasonable trade-off when you get the massive extra sample memory.
The GDS-800 series Goodwills are only 100MS/s for 100MHz bandwidth which is next to useless, you'd have to use interleaved (repetitive) sampling for anything over 20MHz or so.
The only other thing you need to know is that memory depth is just as important as sample rate! I do not know how I ever survived without a deep memory scope.
I'm pro-mixed signal scopes, they are just so versatile, and you don't need a PC to run it. But for home use a PC based logic analyser might be a good compromise, there are at least half a dozen good USB logic analysers on the market. Whichever one you get, make sure it has an external trigger input/output so you can connected it to the scope.
OMG, is that ever sweet! No wonder you don't mess with Rigol much.
Lost in transit...hmm...
places:
formatting link
That sounds like a reasonable rule of thumb.
Sounds like you really go for the deep memory. I assume the scope captures even when it's not triggering so that you can go back and look at the events preceding the trigger. Is that right?
They have some new scopes that do 1GS/s that are replacing the 8xx scopes. Still _only_ 25K points though.
formatting link
Yep, you like the big memory. ;-)
I can't find anything stating that the LogicPort has a trigger out. It samples real fast, so maybe they think you don't need a scope with it. It doesn't have a real deep buffer (2K x 34 pins), but should be fine for most things I can think of doing with it. I can't imagine needing 34 pins.
Correct. All DSO's will default to having the trigger point at half way in the buffer, so you get 50% pre-trigger info and 50% post-trigger info. But you can set it to 100% pre or post data, or anywhere in between.
25K points is still very very useful.
You will too one day, trust me!
The 2K memory is pretty aweful these days, most USB analysers have a lot more - shop around. Deep memory is particually important for digital design.
I have a TDS 1012 and find it a little inconvenient not being able to scroll along (say) an ARINC429 32-bit comms transmission at a reasonable resolution on each bit.
I am curious about why the deep memory is so important to you. This comes up because I have only had a very small number of times where more than the 2k or so on the Tek would have made a difffernce. I do fpga design, general digital design, rf design, if design and low noise low frequency work. I cannot remember ever being unhappy that I did not have a meg of channel memory.
It's not just me, it's almost anyone who uses a DSO, hence the trend to larger memories in even the cheapest of DSO's.
I can't believe you have never needed more than 2K of memory in digital design, that is one area where a large memory is essential, but it you've never had it then you just won't know, you make do with what you've got.
With a scope with only 2K of memory you can only get say 10 cycles on screen and zoom in to see detail (200 points) in each cycle. With a large memory you can turn the timebase right down and capture 5000 cycles yet still zoom in with the same resolution on any cycle. So you can for instance see low frequency modulation on a signal, and then zoom in to see any high frequency glitches in any one of the thousands of cycles.
In digital design a large buffer lets you capture massive amounts of data and zoom in with full detail on say any byte in a 1000 character serial data stream.
For instance, last week I was debugging a GPS NMEA simulator design, and I had to capture several entire NMEA data strings (about 30 characters of RS232 data each) and be able to analyse each bit. Totally impossible with a scope with only 2KB of memory.
Another example from just today, I have been tracking down a digital bug in a third party product that only occurs once every few days or so at best. Data packets get sent once every ms, with a 2MHz data clock, and I want to be able to capture many of these packets, yet I also want to see fine detail in any one of those packets, and I only get ONE shot at it every couple of days. No problem on the large memory scope, simply set it to a slow timebase, it triggers capturing a large amount of packets and then I can spend hours post analysing every packet in detail.
Everyone in the company (and another company who shares our space) comes to borrow my scope all the time because the big memory is just so useful.
Once you have the capability you will never want to go back.
I use the 2K Teks but we have the 20 Meg Lecroy models at work at our EMI testing area, the Lecroy's are almost to the point where you don't need a trigger anymore, just press the single shot button and zoom in. Probably once or twice a year I really need the 20 Meg, mainly to capture independent glitches that occur over a long time period or capturing a serial bus data. Otherwise, I really don't need the Lecroy but they are a huge productivity improvement and a joy to use.
That's really nice. No more probe cables getting hot and letting of that plastics stench. AFAIR they are around $3k, sounds like a pretty good value.
I have nothing against digital scopes as long as their true sample rate is high enough. And as long as they don't spew EMI around, which many of them unfortunately do.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.