Author Topic: Oscilloscope and variable power supply (Read 6518 times)

gamefreak · « **on:** February 16, 2008, 06:13:09 PM »

As I keep working on robots it seems to become ever more obvious that I need a way to get good measurements and test circuits without the fear of breaking them, Can anyone direct me to some good ones that are cheap?

Admin · « **Reply #1 on:** February 17, 2008, 12:30:26 PM »

Oscilloscopes aren't cheap . . . but you can get some low quality ones for like $200+

As for power supplies, you can find a bunch easily with froogle.

gamefreak · « **Reply #2 on:** February 17, 2008, 02:47:26 PM »

any recommendations on a cheap oscilloscope?

Admin · « **Reply #3 on:** February 17, 2008, 03:10:21 PM »

Look for PC based oscilloscopes.

gamefreak · « **Reply #4 on:** February 17, 2008, 03:16:10 PM »

Well there are many different ones, like how many channels and what they can measure, what would be better suited for the determination of timing cycles on a microcontroller?

Also, the first like 4 links all have 1k $ oscilloscopes...

I think that this one: http://www.apogeekits.com/pc_oscilloscope.htm
could work, it has a 50 Mhz frequency, but im not sure....

Admin · « **Reply #5 on:** February 17, 2008, 03:42:01 PM »

Quote

what would be better suited for the determination of timing cycles on a microcontroller?

a simple LED would be best

http://www.societyofrobots.com/microcontroller_time_cycles.shtml

But lets say you wanted to measure certain waveforms . . . well, figure out the frequency you need. Then choose a scope that can handle that frequency.

Look for USB oscilloscope, too.

like this:
http://www.hobbylab.us/

You might also want to try ebay for some old oscopes.

gamefreak · « **Reply #6 on:** February 17, 2008, 04:16:01 PM »

yea I read that tutorial, but the oscilloscope is more accurate and when humans are involved things go wrong.

Also the one you posted only samples at 200 Khz, whereas the atmega 8 and 168 have a default of 8 Mhz, now me and acronyms dont get along, but I think that the Mhz is faster then the Khz.

Admin · « **Reply #7 on:** February 17, 2008, 04:19:46 PM »

Quote

Also the one you posted only samples at 200 Khz, whereas the atmega 8 and 168 have a default of 8 Mhz, now me and acronyms dont get along, but I think that the Mhz is faster then the Khz.

It might run at 8MHz, but you could always program it to output something at 100KHz without a problem. I don't think you care about signals faster than that anyways . . .

paulstreats · « **Reply #8 on:** February 17, 2008, 05:43:05 PM »

farnell sell a bench oscope that works upto 10mhz for something like $160 if i remeber right

Rebelgium · « **Reply #9 on:** February 17, 2008, 05:49:34 PM »

Just my two cents: decent oldfashion oscilloscopes work for years and years. I bought myself a cheap hyoungyiungio oscope (2x20MHz) two years back for 70 euro's, and it still works perfectly

gamefreak · « **Reply #10 on:** February 17, 2008, 06:23:26 PM »

I dont quite understand, if say a sevo requires a pulse between 1 and 2 ms, the oscilloscope would help me determine the correct timing for the micrcontroller, correct? What do you mean by old fashioned? and why would I chose a PC over a desktop oscilloscope?

paulstreats · « **Reply #11 on:** February 17, 2008, 06:31:05 PM »

I cant think of any reason why you would use a pc oscilloscope if you had the choice of getting a bench one. I suppose a pc version has more friendly user interface which is forward upgradeable, but you cant beat an old fashioned scope with a small CRT screen built into it.

An oscilloscope wont let you choose the correct frequency unfortunately, all they do is read in oscillations they dont trigger them out. You can however monitor the pulses coming out of your microcontroller and see if they are correct

gamefreak · « **Reply #12 on:** February 17, 2008, 06:39:44 PM »

They display the length of the pulse in time though, right?

paulstreats · « **Reply #13 on:** February 17, 2008, 06:43:02 PM »

gamefreak · « **Reply #14 on:** February 17, 2008, 06:55:13 PM »

So it then would allow you to set the code to send out the correct pulses.
So Admin says to get PC ones, and you guys say the benchtop ones, Im just looking for the cheapest kind that will allow me to measure the pulses on an 8 to 20 MHz microcontroller. Why would I pick one over the other?

Also, would something like this work? http://www.uni-trend.com/UT81B.html
It is also a multimeter so it is useful for many things

Admin · « **Reply #15 on:** February 20, 2008, 08:08:25 AM »

Quote

why would I chose a PC over a desktop oscilloscope?

The benchtop ones are much better feature-wise, but the PC ones are cheaper and much more portable.
(it sounded like you wanted something cheap and didn't need any advanced features)

If you just want to measure servo pulses (20kHz sufficient), every oscope on the market can handle it.

The item you found appears to be ~$400 . . . what is your budget?

gamefreak · « **Reply #16 on:** February 20, 2008, 04:49:16 PM »

cheapest possible, i found that one on ebay.

Im looking for a nice one for setting up the timing pulses on the microcontrollers, since the Atmega168 has an 8 MHz clock i thought I needed an 8 MHz capable scope.

Admin · « **Reply #17 on:** February 20, 2008, 05:00:49 PM »

Your AVR is running on an 8MHz clock, but you only need to be concerned with the frequency of what you are trying to measure.

For example, your servo pulse is about 1.5ms long (an eternity in microcontroller land).

You want a resolution no larger than 1/10th of that . . . so your scope should take a sample every .15ms minimum. Thats 1/.00015 = 6666.66Hz sample rate.

Therefore a >7kHz sample rate would be sufficient - something any scope can handle.

gamefreak · « **Reply #18 on:** February 20, 2008, 05:58:54 PM »

alright, so if I was going to go off of ebay, then what should I be looking for in an oscilloscope to be used for a long time?

paulstreats · « **Reply #19 on:** February 20, 2008, 06:22:33 PM »

http://export.farnell.com/jsp/Test+Equipment/Oscilloscopes/CALTEK/CS10H/displayProduct.jsp?sku=4283934

gamefreak · « **Reply #20 on:** February 20, 2008, 06:24:28 PM »

that brings me to an export page...with no option of america

So if i had a 150 MHz scope, would it be able to work, or is there such a thing as too fast?

paulstreats · « **Reply #21 on:** February 21, 2008, 06:30:18 AM »

you would buy a probe to handle the difference

gamefreak · « **Reply #22 on:** February 21, 2008, 06:35:40 AM »

See now thats even more confusing, how exactly is that supposed to work then?

paulstreats · « **Reply #23 on:** February 21, 2008, 06:42:31 AM »

I have a 100mhz oscope, and with it there are a selection of probes. There is a 1mhz probe, a 10mhz probe, a 50 mhz probe and a 100mhz probe. The 1mhz probe doesnt work with anything above 1mhz - it does produce some results but they are false readings. The probes are as important as the scope

I think they must have a small capacitor or something built into them.

You can also get probes that measure sound etc (a glorified microphone) that let you see sound waves

gamefreak · « **Reply #24 on:** February 21, 2008, 06:59:17 AM »

alright, so I want 10 Mhz probes then? or would the one ten times as fast work? and can an oscilloscope accept any frequency probe underneath the scopes frequency?

The one on EBAY says its 100 to 349 MHz scope.

News:

Author Topic: Oscilloscope and variable power supply (Read 6518 times)

paulstreats

paulstreats

paulstreats

paulstreats

paulstreats

paulstreats