Oscilloscope recommendations

Hand tools, soldering irons, scopes, multimeters. Talk about em HERE!

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
User avatar
didier
 
Posts: 115
Joined: Mon Nov 17, 2008 6:14 am

Re: Oscilloscope recommendations

Post by didier »

In the early 80's I used a HP1740, as I was a service engineer at ... HP (Former Analytical Division). It was a very reliable tool, I just missed the memory couple of time while I was developing (for my own use) a switching power supply for model trains. It was big and quite heavy but travelled miles and miles without complaining.
I recently restarted electronics in a mixed pro/amateur mode and invested in a RIGOL 1052E: I really want to put a good word on it. It is a smart measurement tool for the price (I bought it really cheap - <300€ - on eBay). 50Mz is enough for microcontrollers, and its sensitivity is convenient for most applications. It has many advanced functions for capturing stochastic or predefined events. Drawbacks? Yes, minor: the cooling fan is a little bit noisy and there is no option for running it on battery.
Comparing both instruments is just like comparing contemporary cars: the RIGOL looks like the latest car I bought for my beloved wife with plenty electronics, memories, driver’s assistance, 300 pages Users Manual (Yes!) and functions that she will probably never use. On the other hand the HP was like one of the good’ol almost indestructible classic cars (like the one under his blanket in my garage!).

HTH

User avatar
rct
 
Posts: 53
Joined: Fri Feb 19, 2010 10:06 am

Re: Oscilloscope recommendations

Post by rct »

Sharing what I've learned so far and hoping to solicit any helpful feedback:

I had been considering getting a USB oscilloscope for price & size reasons. It will be my first scope. I thought I'd wind up having to spend $1k or more for a hardware scope. LadyAda recommends two Tektronix scopes (TDS 1002B 2 channel 60 Mhz b&w $1k) or (TDS 2014B, 4 channel 100 mhz, color for $2k). If I were serious enough about doing enough electronics, I'm sure this would be the right way to go.

The "buying an oscilloscope on ebay" talked me out of getting a used scope off ebay especially as a first purchase.

* http://reviews.ebay.com/Buying-an-Oscil ... 0001568756

Now that I've seen the recommendations for the Rigol DS1052E and price. I think that's the direction I'm going to go. It can be purchased at tequipment.net for $535. Tequipment seems to have good reviews as a seller. Somewhat riskier but cheaper, Dealextreme has it for $405. It is shipped direct from China and may come with chinese manuals only. There are some bad reviews of dealextreme as seller. It looks like it can be difficult to get things resolved with dealextreme if anything good wrong. Still from some of the forums I've seen there are people who get their Rigols from DealExtreme and are very happy.

The rigol is an inexpensive chinese made scope. it has warts, you do seem to get what you pay for, but the general consensus seems to be pretty positive. Here's a video blog review:
* http://www.eevblog.com/2009/04/05/full- ... l-ds1052e/

This one pops the covers and reviews the guts.
* http://www.eevblog.com/2009/10/12/eevbl ... -teardown/

This forum on that site has many pages of posts. Some people have modified the DS1052E which is sold as a 50mhz scope to increase the bandwidth by disabling a filter on the input.
* http://www.eevblog.com/forum/index.php?topic=30.0

There is also a very long running forum on Rigol digital scopes at rcgroups. It's currently 55 pages of posts, starting back in 2007. I found a few of the posts when googling for different bits.

The most consistent complaint is that the fan is too loud. Apparently they've overclocked some chips to get the higher sampling rates at lower prices.

There is windows software called Ultrascope for talking to the scope, but the reviews say it's pretty buggy, almost not worth installing. Hopefully someone will try to reverse engineer the protocol and do something useful for *nix one day. There is support for saving images & csvs to thumb drives or printing to pict bridge printers.

Hope this helps,
--Rob

User avatar
mojo
 
Posts: 136
Joined: Mon Jan 21, 2008 5:04 pm

Re: Oscilloscope recommendations

Post by mojo »

I was going to start a new thread but might as well add to this one.

I am looking to get a digital oscilloscope and am trying to decide between a used Tektronix and a new Chinese one. I am a mostly doing stuff like LED display control and power supplies, up to about 20MHz at the moment. I am moving into doing arcade game systems as well, so need to check video outputs for example.

The Tektronix is an older TDS2022 model. 200MHz at 2GS/sec, 20MHz bandwidth. It's a nice, easy to use 'scope but has two disadvantages: only a 2.5k buffer and no USB. I don't know how much of a problem the small buffer is really as I don't have much experience with DSOs (only analogue), but my usual trick with a logic analyser when looking at a non-repeating protocol is to capture big chunks of data and zoom in to the bit I want. It is also helpful when looking for problems when you are not sure where they are occurring. That isn't possible on the Tek...

Lack of USB is a bit unfortunate too because I find it is helpful to snapshot things for reference when using the LA. It's handy for before/after comparisons and planning/design on the PC.

Another option is the Owon PDS7102 at 100MHz/500Ms/6K which seems to be inferior in every way except that it has a USB port and is a bit cheaper. Then there is the Rigol DS1102E which is 100MHz/1Gs/1M but that drops to 500Ms over two channels. The 1 meg buffer is attractive though, and people generally seem to think that Rigol scopes are pretty good for the money. It is quite a bit cheaper than the Tek too.

The other current contender is a Hantek USB scope. PC only is okay I suppose, although I worry about isolation when dealing with PSUs and video. Very good value for money, except that people say they are not that accurate and the software is a bit fiddly.

All the options are a compromise, but the Tek is attractive because it is a good quality brand which does not compromise on the quality of the measurements nearly as much as the others do. It also "just works", there is no messing around with the innards or 3rd party firmware updates. It is probably overkill for what I want.

Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am

Re: Oscilloscope recommendations

Post by Entropy »

No need to reverse engineer the DS1052E - The SCPI commands are documented in one of the Rigol manuals, and as far as drivers, it's a USB TMC (Test and Measurement Class) device. There's a TMC driver in the mainline Linux kernel and one from Agilent - my experience is that the Agilent driver is far more robust.

There is little to no documentation on the TMC ioctls though, other than the module source. However, one manufacturer of (IIRC) colorimeter equipment posted a reasonably nice HOWTO on USB-TMC under Linux. I don't have the URL to their HOWTO with me at the moment.

I was documenting my progress with using the Rigol in Linux on my blog, but just realized I didn't document my latest discoveries in a post yet... I should do that sometime in the next week or so...

User avatar
rct
 
Posts: 53
Joined: Fri Feb 19, 2010 10:06 am

Re: Oscilloscope recommendations

Post by rct »

Entropy wrote:I was documenting my progress with using the Rigol in Linux on my blog, but just realized I didn't document my latest discoveries in a post yet... I should do that sometime in the next week or so...
Very Cool! where's your blog?

Go ahead, Make me buy that Rigol!

Do you still recommend it? I've been on the fence. It's very tempting. I have to decide between taking a slight chance with Dealextreme for $405, or a slightly safer Tequipment.net for $535.

Thanks.

neik
 
Posts: 6
Joined: Thu Feb 25, 2010 7:28 am

Re: Oscilloscope recommendations

Post by neik »

A couple of comments on linux usb tmc.

The mainline usbtmc driver doesn't seem to be working at the moment. I have much more luck with the driver from:

http://www.home.agilent.com/upload/cmc_ ... sbtmc.html

But it means installing kernel headers and such to get it to compile. There is enough documentation to get stuff working. You can drive a simple device like a multimeter by using SCPI from the command line:

echo "*RST" > /dev/usbtmc1
echo "READ?" > /dev/usbtmc1

Iwrote a simple C program to do data logging from a multimeter. Being able to get stuff onto a computer is pretty much essential these days.

User avatar
rct
 
Posts: 53
Joined: Fri Feb 19, 2010 10:06 am

Re: Oscilloscope recommendations

Post by rct »

I wasn't able to find the blog that Entropy mentioned but I found this discussion on rcgroup.

http://www.rcgroups.com/forums/showthre ... 58&page=12

User wpwrak (Werner) developed a usb tmc driver with libusb for talking to the rigol. It doesn't look like it was every released as a package. I need to slog through the rest of that forum. svn is here:

http://svn.openmoko.org/developers/wern ... /host/tmc/

The driver that handles the Rigol differences is:
http://svn.openmoko.org/developers/wern ... c/usbtmc.c

There are quite a few screen captures here and a few data dumps.
http://people.openmoko.org/werner/rigol/

According to wpwrak, this is a quote useful spec for sending TMC over USB. Note that Rigol's implementation has the little quirk that header and payload are sent in separate USB transfers."
http://www.usb.org/developers/devclass_ ... 1_006a.zip

Some instructions are here:
http://www.rcgroups.com/forums/showpost ... tcount=188

I don't have a rigol scope yet, just trying to slog through the info.

Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am

Re: Oscilloscope recommendations

Post by Entropy »

http://andydodd.blogspot.com/

I am WAY behind on updates though... I made a lot of DS1052E progress after my "Part 2" post but apparently never posted it. I think I may have been waiting until I ported Jeff Mock's Perl-GPIB library to support TMC devices, but I never even really started on that project...

neik has basically summarized most of it. The one thing is that I found a github repo that had a version of the Agilent TMC module with improved build scripts that was pretty easy to install. Don't have the URL at the moment...

I agree with neik that the mainline TMC driver seems broken. It sort of works, but around 50% of commands would fail/timeout for unknown reasons.

The main annoyance with the DS1052E is that when you retrieve a trace from the unit, you get raw ADC readings and not the "converted" voltages.

User avatar
mojo
 
Posts: 136
Joined: Mon Jan 21, 2008 5:04 pm

Re: Oscilloscope recommendations

Post by mojo »

After more research I think the GW Instek 1000A series, and in particular the GDS-1102A 100MHz model, is the best budget 'scope at the moment.

The Rigol is amazing for the money (although don't forget to add import duty), but it seems to need quite a bit of fiddling to make it live up to the claimed specs. The Instek 'scope just works, and according to the reviews it meets or exceeds the stated limits: http://welecw2000a.sourceforge.net/docs ... -1152A.pdf

The spec is pretty good too. 100MHz, 1GS and a massive 2M buffer.

It's not the cheapest but it is cheap, and considerably better than the Rigol and other cheap models which mostly seem to use a Cyclone II FPGA and less shielding etc.

Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am

Re: Oscilloscope recommendations

Post by Entropy »

My Rigol seems to be working just fine without any tweaking...

I didn't have to worry about import duty in my case. I did get nailed with sales tax - had I realized that it would drop-ship directly from the manufacturer (meaning "in-state tax without in-state shipping speed") I probably would have ordered from someone other than Saelig, but not sure. Even with NYS sales tax, Saelig was one of the best prices from a reputable domestic vendor.

sirket
 
Posts: 128
Joined: Fri Oct 26, 2007 8:46 pm

Re: Oscilloscope recommendations

Post by sirket »

mojo wrote:After more research I think the GW Instek 1000A series, and in particular the GDS-1102A 100MHz model, is the best budget 'scope at the moment.

The Rigol is amazing for the money (although don't forget to add import duty), but it seems to need quite a bit of fiddling to make it live up to the claimed specs. The Instek 'scope just works, and according to the reviews it meets or exceeds the stated limits: http://welecw2000a.sourceforge.net/docs ... -1152A.pdf

The spec is pretty good too. 100MHz, 1GS and a massive 2M buffer.

It's not the cheapest but it is cheap, and considerably better than the Rigol and other cheap models which mostly seem to use a Cyclone II FPGA and less shielding etc.
My experience with Instek was _horrendous_.

They shipped me a broken scope- no big deal- stuff gets broken in shipping.

I called them and said it was broken- and asked them to cross ship a replacement (which they could charge to my credit card and then credit when they got the defective scope back) as I needed this right away. Not only wouldn't they cross ship a replacement- but they wanted me to wait a week for it to get to them, a week for them to _repair it_ and then a week to get back to me- on an $1800 scope.

When I said that that was unacceptable the guy told me "I don't care." And then says "it was probably broken in transit- we watch the UPS guys throw stuff into their trucks all the time." I asked if he ever considered asking them to maybe stop throwing the equipment or perhaps packing it better and he responds with "No" and then hung up on me. I documented the store over on the Makezine forums:

http://forums.makezine.com/comments.php ... ionID=2429

Despite all of that- the scope itself was lower quality than I expected. The metal case flexed a lot, it was bigger than a DSO should be, and it just did not have the feel of a Tektronix or even a Rigol scope.

I can't comment on the performance of the GDS 2204, or on the performance of the scope you mention, as I never had a working model to try. That said- customer service is important to me and Instek failed miserably. I only wish I had recorded the call I had with their representative.

User avatar
mojo
 
Posts: 136
Joined: Mon Jan 21, 2008 5:04 pm

Re: Oscilloscope recommendations

Post by mojo »

Sirket, I did read you post on the Make forum. I guess that was Instek in the US. So far I have only emailed their Japanese office but I did get a swift and helpful reply (confirming on specs).

What worries me about Rigol is that they have open-sourced their firmware and seem to expect the community to do their updating for them. Via did that with the C7 and it was a total disaster - the much touted x264 decoding was only ever available in an ancient version of mplayer for Linux and was never usable at all in Windows. It bothers me when companies appear to abandon their hardware like that.

They also appear to have disabled quite a few features because they didn't work, with the promise of re-enabling them in future updates when they are fixed.

Also kind of worrying is all the stuff on Sourceforge for the FPGA. It seems like some very big improvements have been made by open source developers to the code, which makes me wonder what kind of people were employed to write the original stuff if they could not fix what appear to be maths errors.

I'm not saying they are necessarily bad 'scopes, it just seems that they were not very polished, at least at first. Part of me is a big fan of open hardware, for example all my routers run Tomato or DD-WRT, but there is a difference between "open" and "original software isn't that great".

User avatar
rct
 
Posts: 53
Joined: Fri Feb 19, 2010 10:06 am

Re: Oscilloscope recommendations

Post by rct »

mojo wrote:What worries me about Rigol is that they have open-sourced their firmware
That would be very interesting, Can you provide any URLs?
mojo wrote:They also appear to have disabled quite a few features because they didn't work, with the promise of re-enabling them in future updates when they are fixed.
Such as? I'm not challenging what you are saying but I'm trying to make a buying decision, so I'd like what ever specifics I can find. From what I've read, the logic analyzer option for the scope had a bunch of problems. I've recently tried to comb through the 50+ pages at rcgroups and the 14+ pages at eevblog and I didn't really come away with that impression other than the logic analyzer issues. Maybe I didn't understand enough of what I was reading.
mojo wrote:Also kind of worrying is all the stuff on Sourceforge for the FPGA. It seems like some very big improvements have been made by open source developers to the code, which makes me wonder what kind of people were employed to write the original stuff if they could not fix what appear to be maths errors.
My googling didn't turn up anything useful. I'd love to know more.

I may be a total hardware newbie, however, if we're talking in generalities having been employed to work on very proprietary software for a long time on both the vendor and volume end user sides, I don't find this really surprising. What has to be done in commercial efforts to meet deadlines is often full of compromises. Generally good developers after they've delivered a release, will have a list of all of the stuff they'd do differently "next time". Fred Brook's Mythical Man Month from 1975 covered why improvements and bug fixes could always be found. Open source developers, especially if they are 'scratching an itch' aren't typically working with the same constraints (hard deadlines) the way a commercial effort would be.
mojo wrote:Part of me is a big fan of open hardware, for example all my routers run Tomato or DD-WRT, but there is a difference between "open" and "original software isn't that great".
I think you are confusing a number of things here. If the difference between the original proprietary software and more open (i.e. hackable) alternatives "isn't that great", why would anyone use DD-WRT, Tomato, OpenWRT? Why are you bothering to void your warranty and install one of those?

Second, The broadcom, atheros, etc. hardware that DD-wrt, tomato, etc. run on could hardly be called open hardware. You are confusing open hardware with open source software. Broadcom & atheros may have reference platforms for their chipsets that manufacturers can easily turn into products, but that doesn't make them open hardware. Look at why many of those are still running a linux 2.4 based kernel, due to lack of open software and/or specifications. Look at the b43 wireless drivers for example. They need to pull the broadcom binary bits out of the windows driver packages in order to be able to deliver you a working setup on an open source system. Many open source developers are trying to reverse engineer the software, firmware etc, so they can write free software without having to sign an NDA in order to get chip specs that would then encumber their activities.

In any case, I'd really like to hear any specifics you can point to about the Rigol DS1052E. I'm sure it will help me & others make buying decisions. I really do want to hear any negatives.

Thanks.

Entropy
 
Posts: 472
Joined: Tue Jan 08, 2008 12:43 am

Re: Oscilloscope recommendations

Post by Entropy »

mojo wrote:Sirket, I did read you post on the Make forum. I guess that was Instek in the US. So far I have only emailed their Japanese office but I did get a swift and helpful reply (confirming on specs).

What worries me about Rigol is that they have open-sourced their firmware and seem to expect the community to do their updating for them. Via did that with the C7 and it was a total disaster - the much touted x264 decoding was only ever available in an ancient version of mplayer for Linux and was never usable at all in Windows. It bothers me when companies appear to abandon their hardware like that.

They also appear to have disabled quite a few features because they didn't work, with the promise of re-enabling them in future updates when they are fixed.

Also kind of worrying is all the stuff on Sourceforge for the FPGA. It seems like some very big improvements have been made by open source developers to the code, which makes me wonder what kind of people were employed to write the original stuff if they could not fix what appear to be maths errors.

I'm not saying they are necessarily bad 'scopes, it just seems that they were not very polished, at least at first. Part of me is a big fan of open hardware, for example all my routers run Tomato or DD-WRT, but there is a difference between "open" and "original software isn't that great".
[citation needed]

Simple as that. You don't have a single link to back up any of your claims, and at least so far, attempts at Googling your claims turn up nothing.

I haven't seen anything indicating the Rigol firmware is open source. The closest thing I've found so far is a comment in EEVblog that another scope manufacturer in Germany went bankrupt and open sourced their firmware.

sirket
 
Posts: 128
Joined: Fri Oct 26, 2007 8:46 pm

Re: Oscilloscope recommendations

Post by sirket »

mojo wrote:Sirket, I did read you post on the Make forum. I guess that was Instek in the US. So far I have only emailed their Japanese office but I did get a swift and helpful reply (confirming on specs).
My apologies- I was not trying to say the Rigol scopes were better than the Instek- as I said- I never got to try one. I did get to see the display (though no waveform showed up) and I did get to try the menus. It seemed nice enough although- as I said- the GDS2204 is a much bigger scope than the equivalent Tektronix. The model you are referring to seems to be much closer to the Tek in size.

As I'm based in the US- I dealt with their US customer service and it was so bad I just won't deal with them again. I ended up buying a Tektronix TDS2024B. It was a little more expensive- but the customer service was _amazing_. The pouches for the scope probes that I received were all cracking and breaking down. I emailed them about it. Within an hour someone had gotten back to me- asked for permission to call me. They called and told me they were sending out new pouches immediately- but that they also wanted to know more about the defective pouches so they could make sure it didn't happen again. Then they followed up with me to make sure I received the new pouches and to let me know that they were going to monitor their supplier more closely to ensure it did not happen again. All of this for $2 worth of probe pouches.
Last edited by sirket on Fri Mar 12, 2010 1:17 pm, edited 1 time in total.

Locked
Please be positive and constructive with your questions and comments.

Return to “Tools Tools Tools”