What ocilloscope should I get?
boeboy
Posts: 301
first of all·I am going to get my first ocilloscope·I need a bench top one and·I·need·to know how meny·MHz·I need·for use with the·prop i·am leaning towards a 20 MHz·ocilloscope but I do not know if it will be adequate so if some one out there has one and will share his\her expertise.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
Post Edited (boeboy) : 2/23/2007 2:56:59 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
Post Edited (boeboy) : 2/23/2007 2:56:59 PM GMT
Comments
For openers, what kind of signals do you want to look at?
Audio signals, or any signal less than 40khz with slow rise and fall times,·can be studied on a PC's audio card. There are several downloadable programs that will allow you to do this. You'll have to make probes for this job. Seems to me I've seen one or two recent threads here about that, so do a search. You can probably use this to study automotive circuits, but it will miss most of the troublesome noise that screws with TTL circuits and processors.
If you plan to study clock signals, or any square wave, multiply the clock's frequency times 10 to get the needed o-scope's bandwidth rating. So, a 10 MHz clock (or most any processor output) needs 10MHz times 10, or a 100MHz o-scope, to do the job properly. By properly I mean being able to study the rise and fall times of each signal without degradation. Put it this way: A 10MHz clock on a 10 MHz-rated o-scope will look like a sine wave. Stating it another way, a 60 MHz o-scope will reliably track a square wave up to 60 MHz/10, or 6 MHz.
If you want to save the signals for later viewing or analysis, you'll want a DSO (Digital Sampling Oscilloscope). These get expensive real quick. A decent 60 MHz model will get above $1000. These introduce another factor: sampling rate. Nyquist's theorem suggests that the sampling rate be more than twice as fast as the signal you're analyzing. In fact, twice+1 sample would suffice, theoretically. So, if you want to study a 20 MHz· clock, your sampling sampling rate should be 40 Million (+ 1)·samples/second. Of course, you'll need a 200 MHz bandwidth, as discussed above. Other important factors·are the storage depth (how much memory per channel), how storage is distributed, the ADC's (Analog-to-Digital Converter) bit count ( 8 is typical, more is better), how it connects to the PC (not always available; USB is best, serial is OK), and a raft of others.
Check out Ebay. There are lots of o-scopes going for less then $300. A lot are old workhouse Tektronix models, most doing 100 MHz. Most are analog models, so no storage is possible, but there are a few DSOs available. There have been a few slower models (2, 5, 10) MHz o-scopes for under $500. These include dual-channel handhelds with builtin DMMs (Digital MultiMeters).
No doubt you'll get lots of useful info here. These forums have a HUGE body of really knowledgeable people.
Good luck in your search,
kenjj
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
I just checked Ebay doing a search using "100MHz oscilloscope". There are plenty of models available, some complete, several from $50 to $200 (or more). There's even a few four channel models, one for $350. These units come in various states of repair, and various amounts of manuals and probes and accessories. Just about every manufacturer is represented. There are even a few workhorse Tektronix 465s (purely analog) in there. Probe sets of four or more 100MHz probes for HP and/or Tektronix are offered, just in case a scope catches your fancy but comes with no probes.
I imagine most are uncalibrated, offered "AS-IS", "FOR PARTS", and of unknown usability. Read the descriptions then come back here with any questions that you have about features and such. If you see a scope with a users' history saying it was on his bench, or has a recent calibration label, that's the one to go for. You may have to pay more, but the confidence it will work out of the box is really worth it.
Also, watch out for shipping charges. Some of these auctioneers price cheap but charge dearly for shipping. That way when you scream it doesn't work and you want a refund, he of course doesn't refund any of the shipping and so pockets the difference. And you get stuck paying shipping to send it back to get your refund. Read the description VERY carefully for warranty and fitness for use and refund processes. BTW, if shipping insurance is offered, take it. If not, see if you can get it. If the shipping company screws the pooch they will only offer some pitiful default amount as compensation, if any.
One of Ebay's saving graces is that it offers some guarantee of some amount of money back to you if you win the auction and the seller screws you in some manner: failure to ship, shipping something other then what appeared in the auction, equipment Dead On Arrival, etc. They offer a mediation service between the seller and buyer, too.
I feel that a 100MHz analog scope will do you want based on your using it with BS2s and such. Two channels is typical, four is better. Digital sampling is nice, but overkill if all you want to do is check for activity on a pin. For that matter, a $25-$40 logic probe (brand new) will do that much!
Check it out. And have fun doing it!
kenjj
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
I would probably opt for the one that comes with the text... http://www.parallax.com/detail.asp?product_id=28119
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
NerdMaster
For
Life
Well - my opinion....
Check out EBay for the Tektronix 7000 series - I bought a 7603 for $99.00 and then 3 plug-ins for another $99.00
One is an amp (vertical), one is a counter (225Mhz) and one is a time-base.
Now if you move 'up' to the 79** series, you can buy an amp that'll do 1Ghz! - of course probes that'll perform that fast
are a tough item to find....
Point is...in the 1970's these scopes were top of the line, now there's an ultra-reliable scope on my workbench that
was just out of calibration - it let's me look at some very fast signals....clock pins-I/O pins....
and it's a storage scope - I don't think you can afford one of these new DSO's - I know I can't.....
Tektronix has always made some of the worlds finest scopes, HP's are nothing to scoff at either!
LeCroy made some very exotic scopes - they sell cheap on EBay because nobody has heard of them.
Shop Wavetek for function generators, good deals abound....you can get some heavy duty (aged)
equipment for very reasonable!
Just my 2 cents..... or 2 LSB's worth!
John
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
lets see what this does... KA BOOM (note to self do not cross red and black)
With the death of my Grandfather, I inherited a number of interesting things; one thing is a Tek 422, sans probes and manual.
It's filthy on the outside, but nothing is obviously broken. It's got the AC adapter, it powers up, and as I understand the "paperclip test" (actually done with a short piece of hookup wire), it's working and showing a nice sine wave. All of the adjustments seem to work.
So, my question is, not knowing anything about o-scopes, is it worth trying to find some budget probes, download the manual from bama.sbc.edu, and learn as much as I can about using a scope before sinking the "big bucks" on a newer one, or is the 15MHz bandwidth too limiting to make it worthwhile?
Thanks
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer
Parallax, Inc.
Post Edited (Paul Baker (Parallax)) : 2/27/2007 6:03:23 AM GMT
Well, I visited Ebay again.
Searching on "TEK 422" found Item number: 270092001328. This is a cleaned and calibrated 422, with a fairly decent list of specifications. The specs suggest that your 'scope will work just fine with Stamps and such. It should easily handle serial communications, up to the upper limits, like 115200 Baud. It's going for >$100, with a 30 day trial period. You have to calculate shipping, but he does offer insurance. This needs probes, however, which brings us to...
Item number: 250088919446, a pair of 100MHz probes billed as suitable for HP or TEK, for just $20, with $9 shipping. These should do you just fine.
And, as for manuals and such, we have Item number: 130080635188, a TEK manual on CD for the 422. Says brand new, and looks it. Shipping is $4.79 in the US.
Hope this helps.
kenjj
Apologies.
kenjj
Nice acquisition! - If I remember correctly the 422 was considered 'portable' (hhhmm, heavy, but portable)...
If it has the battery pack, for goodness sake, replace those batteries! - they're old and yes old NiCads will leak!
YES! - your o'scope will do most of what you want it to do! - 15Mhz will allow you to 'look' at many signals!
The 422 is an excellent 'portable' scope! - it should give you many more years of good service!
The only part I wouldn't agree with is it is 'portable' - they're heavier than heck!
You can get simple oscilloscope probes on EBay or from Jameco, You're looking for a simple 100Mhz probe....
There's a little switch that says 1X and 10X on them, they are compensated for higher frequencies....
Switch it on 1X for starters.....
I'll 'go off' on a little RANT, if I may.... I believe an oscilloscope is a very valuable piece of gear - because you can't see
electricity! - you have to 'trust' books and manuals to tell you things like 'on the leading edge of a signal' - what?
Trust? - well, with an oscilloscope you can see things... you can troubleshoot many problems and well, there are many
ways you can detect if signals are happening - most people have visual learning abilities, you see it and it makes sense!
If you need to know what a 'pin' is doing on your board....you can just look! - make sure the probe is 'grounded' - this is easy on
ALL parallax boards! - the little mounting holes around the edge of the board - convenient.
On the parallax propellor Demo board they give you this HUGE ground lug! - NICE!
Your 'probe' should have a nice little ping grabber to hook onto the leads of LED's and parts.....
Careful not to short anything out! - go ahead and just 'poke' around a circuit....
There'll be a switch next to the probe that says AC - GND - DC....
Switch it to GND to begin with, use the vertical position knob to 'center the line' in the oscope screen...
then, switch it to either AC or DC coupling - if you're poking around in TTL or logic circuits, switch it to DC!
Then, use the Vertical scale knob to get a 'scale' that is useable! - usually 2V for TTL 'stuff' -
That's 2V per division on the screen graticule....
Now, there's a timebase - it's how long the 'trace' stays repeats itself on the screen....
To the left will let you watch very slow moving signals...
To the right will let you watch very fast moving signals....
Turn it until you see something that makes sense!
There are 'delays' on triggering on some scopes, turn that 'off'...
Now, if everything is set just right, you should be able to view things like:
Serial signals up to 115200 baud - just like someone suggested in this thread!
Input signals from other 'chips'
Output signal states...
I know this is a long rant, but I believe an oscope is a valuable tool
to learn about electronics and I really believe everyone needs a stand alone scope...
I don't see many of them any more.... Everyone has a computer....
And frankly, a PC based scope isn't my favorite... it ties up a computer...
Scopes aren't very useful for RF though - very few of them are 'fast enough'....
But for digital 'stuff' - just my 2 LSB's - they're indispensable, priceless and a vanishing item
on 'the bench'....
Just twiddling my bits....
John