Oscilloscope for embbeded syetem.

Hi,

I have to start working on two boards

TS-7200 and an fpga

I was wondering what kind on oscilloscope should I get?

With TS-7200, I am going to be using pc like protocols(rs323--i2c), over general purpose pins.

Thank you.

--------------------------------------- Posted through

formatting link

Reply to
aladin
Loading thread data ...

If you'll only be debugging low-speed stuff like RS232, you can pick up a $40 pocket scope off eBay and it will be more than enough. There's a bit of a spectrum between cost and convenience when you're talking real scopes for real work, though - for instance, you'll probably be able to do a lot of your debugging on the FPGA side of the board with a 2- or 4-channel analog scope of say 100MHz bandwidth

I have an old Tektronix TDS210 60MHz digital 2-channel scope (inexpensive on eBay now) on which I do practically all of my debugging. I also own, but unwisely loaned out, an HP 54645D scope with 2 analog channels and 16 digital channels - a mini logic analyzer

- which is MUCH more useful. If you can get it, a scope like that makes your life very much easier because of the digital channels, and also because the HP scopes have great memory depth - capture a signal zoomed out, and zoom in after you've frozen it to see the detail. Saves you a lot of time tinkering with trigger conditions to capture a particular event.

Reply to
larwe

Sounds a little like one of these

formatting link
I've been trying to convince myself that I don't really need to get yet another toy^H^H^H piece of test gear. It does seem to be one of the few USB 'scopes out there that actually has a decent sample rate.

--
Rich Webb     Norfolk, VA
Reply to
Rich Webb

I am very reluctant to buy significant-cost PC-tethered instrumentation that can't be used standalone, though I have been tempted a few times (I2C/SPI grabber tools for instance). In the time I've owned my TDS210 I've been through OS/2 Warp through Windows 7 (with Linux and MacOS variants). Computers and TE just don't have the same lifecycle...

Reply to
larwe

USB tethering of instruments is great if you have an on the road requirement, or if flying with the scope a lot as well.

However most scopes these days are not as big as they used to be with LCD screen meaning smaller cases as the need for high voltages and long tubes has gone away.

Could not agree more with the lifecycle as

1/ My bench has enough USB leads and other bits already, limits distance between scope and PC. In otherwords potentially clutters bench with PC(s).

2/ When working with kit that needs to talk to PC I have enough windows open already without adding more confusing things. Worst case you end up with two PCs due to some driver or OS mismatch.

3/ Occasionally you only really need to grab screenshots or waveform data for other use on PC. Unless it is a production test and qualification requirement that has to save lots of waveforms for each device under test, you often use the software to control the USB scope and display the image.

4/ If you end up doing stuff faster than about 10MHz, watch for scopes that do repetitive sampling, often I want a particular event trigger like 25MHz SPI data stream particular byte or packet, pixel event on a specific line of an image.

5/ Modern scopes often have abilities to trigger and decode on many busses and signal formats from video lines, to I2C address/conditions SPI bus activity, or even UART data. This can often be decoded on screen for you, saving a lot of time. The triggering on patterns means you can get to events quicker.

6/ Most scopes these days have on screen measurement capabilities see what is important to you.

7/ Ethernet connectivity is also possible for scopes meaning easier to grab lots of data to where ever.

I ended up last year upgrading to a Tektronix MSO2024, 4 channels @ 200MHz BW, 16 digital channels, 1M sample storage depth with zooming 7inch VGA LCD I2C/SPI/UART decode/trigger USB storage (Type A) and USB control (Type B) connectors Ethernet connectivity VGA output for external monitor if required.

If doing FPGA type work the digital channels will be a big bonus. I went for 4 analog channels for those times when you are trying to capture an SPI transaction and check for ringing, glitches, incorrect levels.

The BIG caveat is how much money you have to spend on a scope, but look at it as an investment for longer lifespan than any PC.

I eventually got rid of a Tek475 scope after nearly 12 years of service other than stored PCs, I had changed PCs three times in that time.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Timing Diagram Font
  GNU H8 - compiler & Renesas H8/H8S/H8 Tiny
 For those web sites you hate
Reply to
Paul

Used to exclusively use a Tektronix 254. Sadly that died a while ago so now use the TPS2024. This has four isolated channels (useful for fault finding motor power drive circuits) and has 200MHz bandwidth, is totally portable and very lightweight to carry, and features a reasonable set of capture and analysis functions.

--
********************************************************************
Paul E. Bennett...............
Forth based HIDECS Consultancy
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-510979
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
********************************************************************
Reply to
Paul E. Bennett

"aladin" wrote

I also suggest a good mixed signal scope for convenient handling. Look also at the Hameg HMO series, but currently the firmware has a flaw displaying digital channels with a time offset in Zoom mode.

A TechTools Digiview logic analyzer is also nice (in addition to a "normal" scope) and worth the money (1/10 the amount of a MSO).

Oliver

--
Oliver Betz, Munich
despammed.com is broken, use Reply-To:
Reply to
Oliver Betz

I love my Agilent 54622D. I bought it when I started consulting, and it's paid for itself several times over on embedded projects. Way spendy -- if there's a budget minded alternative now you may want to go for it, but if you're working for bux it's a sensible buy.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html
Reply to
Tim Wescott

To add another piece to the puzzle, some low-end HP - sorry, Agilent - scopes are re-branded Rigol units. (Chinese) They can be bought cheaper under the original brand.

Then:

formatting link

or (Tekway)

formatting link

-- Roberto Waltman

Reply to
Roberto Waltman

I got my TDS210 way back in the late 90s when I first started as a self-employed developer. I still use it now. When I got it, my wife doubted the wisdom of spending $1000 on a piece of test gear as it was a significant chunk of my income in the first lean years. On a cost per hour used, it has to be one of my all-time best buys, though.

Since I do most of my work with processors having clocks under 20MHz, scope speed hasn't been much of an issue. A set of digital channels would be nice, as would more capture memory. However, I'm reluctant to go out and buy a newer scope just for those things.

Mark Borgerson

Reply to
Mark Borgerson

That is way cool. Now I'm looking at my old TDS210 and thinking "I could upgrade you to a 4 channel color unit...."

Reply to
larwe

Watch for when repetitive sample kicks in. One of the reasons I discounted HP/Rigol and a few others.

Been caught by that many times on other people's scopes.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Timing Diagram Font
  GNU H8 - compiler & Renesas H8/H8S/H8 Tiny
 For those web sites you hate
Reply to
Paul

I spent a lot of time checking specs on and off over a year before getting to a short list and got a good deal on an ex-demo unit.

Sometimes it gets time to upgrade units.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Timing Diagram Font
  GNU H8 - compiler & Renesas H8/H8S/H8 Tiny
 For those web sites you hate
Reply to
Paul

One thing worth considering is a good quality analogue scope, as it's a lower cost alternative if you need high speed capability. Having said that, modern digital scopes are pretty good and the downside of analogue is the lack of storage. They both have their niche applications, so one of each isn't a bad idea. I started out with s/hand solartron, tektronix

465 and similar, but now have a 2465 tektronix analog and hp 54600a + a 54701A for flexibility. Have had the 2465 for around 7 years now and had to repair the switch mode psu last year, but it's probably 15 years old now. Tube is still good and bright though and there are plenty of spares around on Ebay. Also have an HP 16500B logic analyser / scope mainframe, which can take a wide variety of analyser. scope and word generator plugins. It's big and clunky, but it can be found at firesale prices and has a lot more capability than a dedicated scope or logic analyser alone. Mine cost
Reply to
ChrisQ

That brings back some memories. In 1983 I bought a brand new 2235A. Tek financed it, which caused me to give them my lifetime loyalty.

It's still on my bench, still has a bright trace.

Same issue with the wife too...

Reply to
Jim Stewart

So I'm hearing that your wife still has good bright phosphor and doesn't need recapping? Impressive!

Reply to
larwe

For the braindead (remember I officially work in Marketing now, not Engineering ;)) what does this actually mean? Are you referring to the fact that (at least on my erstwhile HP) when you zoom in, at some point it starts interpolating samples that aren't there by drawing vectors between real points?

Reply to
larwe

Thats right, I've got my DSO5102B, reflashed it to DSO5202B and now have a

200MHz BW scope for a 100MHz price. Currently using to debug my design with Spartan-3E board and I am VERY happy with it! All the instructions about mod are on the eevblog forum and my blog :)

formatting link

Reply to
scrts

Well, I don't know about you guys, but my wife tops the TDS210 significantly in return on investment! ;-)

Mark Borgerson

Reply to
Mark Borgerson

Beyond a certain sampling speed (not timebase setting), the scope takes samples of the waveform many times slightly offset to build up a higher effective sample rate of the waveform. this is good for looking at continuous waveforms (sine wave), but lousy for looking at high speed datastreams or digitial video down to pixels or their edges, as these change every time it builds up the waveform samples, usually for each triggered event. So looking for a particular packet in a datastream or the pixel on a line can be difficult, as either the event happens once and their is net to no samples to display anything useful, or you cannot guarantee what you are seeing is valid.

Your datasheet has to be read very carfully. A usual giveaway is terms like 'Efective Sampling rate'.

I know Rigol/HP do this and I have seen this on Yamamachi scopes at what I would consider a fairly low sampling rate.

Often this is NOT something you KNOWINGLY choose, and you need to know what the actual max sampling rate is, which is different to bandwidth of scope. You would be amazed how many lower cost upto 200MHz scopes actually have sampling rates of the order of 100 to 500 Mhz and use repetitive sampling to acheive the EFFECTIVE 1 to 2 Gsa/s.

Repetitive sampling goes under many (mainly marketing) phrases including Effective Time Sampling.

You need to know on a scope when this will happen, as this is built into the scope and is NOT A USER SETTING.

See

formatting link

14295&lc=EN

formatting link

A Tek primer on it.

formatting link
df

formatting link

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Timing Diagram Font
  GNU H8 - compiler & Renesas H8/H8S/H8 Tiny
 For those web sites you hate
Reply to
Paul

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.