EEVblog #86 – Buy a real Analog Oscilloscope PLEASE!Posted on May 11th, 2010 59 comments
Dave is sick of getting emails asking if the cheap toy DSO’s and FPGA/PIC/AVR oscilloscope kits are worth buying.
Guess his response… answers by telepathy please.Drive Time, EEVblog - Official Releases, EEVblog - Podcast digital, fpga, measurement, oscilloscope, rant
Yep, I have a Tektronix 465. The Tek is a bit of a boat anchor.
And yes, I didn’t pay a dime for the 465. It was a ‘gift’ from a friend.
However if you could give me a 20MHz scope that was pocket size I’d be very hard pressed not to get one. But my next will probably be the Rigol DS1052E.
You should do a full singing blog
Agreed 100%. so often do i go in the lab and watch people push the damn auto set button and not truly know whats going on. The first scope I used and was a tektronix 2336. some of my profs at university hate the auto set button about as much as I do, never shows me what I want. Wish I didn’t have to sell my old scope but I have to pay for my new power supply.
Previous vblog on DSOs was EEVBlog #12:
2/2 should be: http://www.youtube.com/watch?v=Ev121xAt_k4
if anyone has an anlog scope and wants to support our local hackerspace it would be very much apprechiated by our members. only problem is it needs to come to bangkok somehow
forgot to mention that was a very good advice, was eyeing with both the cheapo scopes and a rigol, but there are tons of other things i would prefer to put cash into.
Ppl who are wondering to get “scopes” like DSOnano obv. never used “real” scope. I have DSO nano and Rigol DS1052E (100MHz mod ftw) and i’d never want some kind of toy scope, even 10 years old analog one have more possibilities than those ones. And PC-based scopes… i dont wanna burn my $1000 PC coz some $100 dollar PC scope don’t have proper input protection and separation.
Like most things – if you’re serious buy the best you can afford. For me it was a secondhand Tek2465, short of the latest digital ‘scopes, its hard to beat. They seem to be pretty easy to get these days via Ebay, but you’d want to be pretty skilled to repair one – there’s black magic inside! Once you’ve used good gear, its hard to go backwards.
Having said that, for low frequency stuff, the humble PC sound card can be pressed into service – better than nothing and is quite handy for doing FFTs and recording IR remote control data streams. Best to use a usb one (can be got for a couple of $$$) in case you do something wrong and zap it. Also good for a cheap ‘n’ cheerful signal generator.
Dave – have you done a blog on soldering irons yet? I usually don’t touch anything below a Pace, so I’m out of touch with the low end!
Looking at your forum, you have acquired a good number of trash-talkers, don’t you? I mean the people advocating stuff like these toy ‘oscilloscopes’ and dangerous $5 ‘multimeters’.
I cringed when I read that trash about analog oscilloscopes being difficult to use. Where do these self-styled EEs come from? From the University of Eternal Delusion?
The function (usually a menu item) I use most on a DSO is reset, not the auto set button. I want the oscilloscope to be in a known state before I start, not in a random state guessed by some software, written by some guy under time pressure. DSOs have all sorts of hidden settings, while good old analog ones just require to scan the front-panel and fiddle with the knobs. Left-to-right, top-to-bottom, “do I need this feature, what value should this have?”, and a quick press of a button or turn of a knob and I am in business.
At least one company, Goodwill, sells DSOs where the auto set button can be locked by teachers or instructors. Unfortunately, they also published the unlocking code.
>Looking at your forum, you have acquired a good number of trash-talkers, don
I haven’t registered for the forum because of them.
Are you suggesting some of the “Hero Members” should leave? if yes, then you should have a serious discussion in PM with dave, IMO.
and btw, where are most of all the EE Pro’s should be at right now? googling, foruming, chatting online? if that’s the person we are trying to advertise to, then for sure, Youtube and EEVBlog are not the suitable place to do. Somewhere in factory, mining coal or Open Sea Oil Tanker Engineer that shuld be told of this.
do you have your own usefull and informative site to suggest pal?
Yep, I have an ancient Tek 7704A modular scope with four channels at 200 MHz, dual timebase and digital readouts. For the princely sum $150. It works great and spare modules are usually available on ebay. When this beast came out it cost more than I made in a year as an engineer. Fortunately I have the room for it. Modern DSOs are great but this thing works for most of my needs.
Hello! You are so right, an old analog scope is better than a cheap USB DSO. For three years I had an old 20 MHz, 2 channel analog Metrix, got it for free from a lab where my father was working. It was really usefull, I was even able to repair a plasma TV with it; I learned a lot about using an oscilloscope, the input, the bandwidth. It was not sufficient for digital electronics and microcontrollers, so I was going to buy a second-hand digital Tek a few weeks ago, but then I learned about the Rigol DS1052E, I purchased it and I am very happy with it now!
The feeling is a little bit different with digital scopes, you always have this little ADC noise (even if they try to correct it), and you feel that you’re not seeing all your signal, as it’s only recorded 20 ou 30 times per second to give you a smooth image… So I think I’ll still be using an analog scope in some cases (like audio equipement).
I got a Tek 475 (identical with the 465 but
wider bandwidth) then found a 485 dirt
cheap at a Hamfest. HP also has very good
old scopes out there but they cost a bit more
(go figure). (I worked at HP (Agilent)
in the 60′s: best test equip, but Tek is
the leader in scopes).
The 465-485′s are 70′s vintage so still
fixable by an amateur. Later models
are all special IC’s that no longer exist.
In fact, the old 70′s test equipment from
Tektronix and HP is great stuff: accurate,
fixable, and dependable. (and big and heavy)
I have a Bitscope but it is harder to use
and very limited. Your advice is right
on. A good DSO, in my opinion, is much
harder to use and best to get after you really
learn how to use a scope and do it often.
DSO’s do more, but if you can’t use it, who
Oh, from May to December apparently Rigol
has the DS1052D on sale for $399. Check
Fair comments. It’s not fair to tarnish all USB scopes as rubbish though I have a Picoscope
And it’s far from a rubbish toy saves a lot of real estate too and I have almost limitless storage via the PC. Mind they are not cheap.
and Yup I started with an analogue scope , mainly because back in the day there was no such thing as a DSO it was analogue or bust baby.
What about the other cheap *real* scopes beside Rigols? I mean Atten, Owon, Uni-T, Welec/Wittig.
So far googling around it looks like:
-Rigols are the best bang for the buck.
-Attens mostly are lower quality Rigol knock offs.
-Owons are generally cheap mediocre quality scopes.
-I have no data about the Uni-T (comments welcome!).
-Welec/Wittig have a defective firmware and other issues that severely limits their usage. Hence their low price on Ebay.
Comments/corrections welcome. I especially would like to get some first hand impressions about the Uni-T scopes.
“Comments/corrections welcome. I especially would like to get some first hand impressions about the Uni-T scopes.”
I would not recommend the uni-t scopes because of the horrible PC control software (for the 2000 & 3000 series atleast). It’s built by someone just learning LabWindows/CVI, the released versions is built as debug, they haven’t bothered to switch from the default target or remove the _dbg suffix from the executable name!. It’s built on an ancient version of CVI (8.0), the graph control is not even accelerated, and crashes when it feels like it.
They’ve actually updated the software a few times, but mostly just changing the colors of the rotary controls and fixing some spelling errors (still plenty left) There is no NI Instrument driver or other library available which would have been useful for your own projects, and of course there is no driver for Vista/Win7. (The Chinese version is a little bit better but still bad)
I’m thinking of writing them asking for a IVI driver or some code (good luck?).. Or perhaps writing my own as a learning project this summer(my scope uses cypress ez-usb fx2 for USB connectivity). Since software has debug symbols, and uses documented cypress driver (which is available for vista/win7, yey) maybe it’s not that hard.. some other frustrated uni-t owner wanna join me?
Otherwise as a stand-alone scope, I’ve had my UT2000-series for 1-2 years now but only done basic hobbyist stuff, and I’m no pro scope reviewer can’t really comment. It was only $300 or so, but the trouble is not worth it =)
I agree 100%
We say the auto button as the “grandma button”.
I bought my 20MHz analog scope for 15
Do you by any chance have studied at the university of applied sciences in D
Wow! Witch song was that! Who is the singer? Thank you dave ! =)
Just bought a digital scope (the Rigol coincidentally), but I still very much miss having an analog scope. I am only 24 years old, so when I went to school digital was just becoming big and affordable. The experience I had from us an analog scope put me ahead of a lot of people who just used the auto button from day one.
I use the auto button on very simple waveforms, and usually end up tweaking controls a bit to get what I want.
I am also still amazed by the number of people who don’t properly understand and know how to use various triggering modes…so frustrating to watch sometimes.
I think everyone should have to start on an analog scope.
Got myself a used HP 54645 MSO (2 Channels 100Mhz+16 channels LA) off ebay for $300. I did have to change the NVRAM chip, whose battery died in the last 12 years. But now I have a high end MSO for not a lot of money
And these old instruments are not only built like tanks, they are also very user friendly.(I have Rigol DS scopes at work for comparison)
I think key to getting a USED good brand analog oscilloscope is you need to anticipate fixing it, as a few others have said. Most are over 30 years old, and newer ones have increasingly specialized parts. At least, many of its internal electrolytic caps are past their predicted lifespan. So while it makes good bang for buck, I’m not sure a complete newbie will be happy buying a second hand scope that, knowing nothing about scopes, will not know the differences between a malfunctioning one and one that works as it should.
That’s actually a really good point… I’m a hobbyist, and I previously had 2 analog eBay clunkers, and I had serious trouble knowing (1) whether they were working properly, and (2) if they were uncalibrated, what negative affect that would have on my results.
I ended up getting this scope recently:
Dave, I’d guess you would say this scope is probably garbage, but it’s been really helpful for me– I’ve been able to display video signals and oscillator output and identify the parts of different signals. So in general, I’d have to say it’s been much more useful than the other 2 eBay scopes I’ve had.
Dave, what do you recommend as far as calibration?
No I’m from Scandinavia. And yes, somepeople call it as the “grandma button” around here.
Sometimes I accidentally hit the auto button on my Rigol, which resets all settings to an unpredictable state. I wish there would be an option to disable the button completely. Or even more: alter the function of this button to instead instantly enable the “single” triggering mode.
BTW, does anyone need a Tektronix 453 scope with probes for free in Seattle area? Works OK, although I suspect the triggering should work better… Google it to see what it looks like.
The thing is so heavy to ship. Pick up @ Sammamish. Email me tk19900 () yahoo dot com if interested, before I put it on craiglist. I think I have the schematics and manuals in PDFs.
tchicago, if you want to disable the auto button, just pop it out.
Do You still have the 553 ?
I got a B+K Precision 1479B (Dual channel 30MHz analog scope) for 12USD from my school’s quarterly surplus sale about a year or so ago, and it’s a great scope, though the astig control has a bad cap somewhere and I’ll need an ESR meter to find it without the damn overpriced repair manual.
Anyways, 30MHz is a useful amount of bandwidth for most applications, I can see common AM and FM IF frequencies and microcontroller PWM, so it’s pretty nice.
I have a Tek 7834 (4 plugin 400 Mhz analog storage) w/ 7B92A (horiz delaying timebase), 7a26 / 7a13 / 7A18 vertical plugins and a 7L5 w/option 11 (L3 input impedance module) and option 25 (tracking generator). I know… boatanchors, but I fell in love with this scope when I worked at Hughes Aircraft, and swore that one day…..
oopss…. the 7l5 is a spectrum analyzer 0 to 5 Mhz
Also JUST bought a Rigol ds1052E (and did the firmware mod thanks to Dave!) NICE SCOPE!!!!
i bought the dso nano. i knew in before it is not a replacement for a real oscilloscope, but the hardware and software is open source and that is worth supporting it, in my option.
and if anyone says this is a complete bullshit and not worth spending a dime, i answer: make a better open source one!
overwriting the serialnumber for gaining more features … i call that bullshit!
Hi I do not agree with you in some way about the little DSO Nano scoop I have sold many of them and all where very happy with it! But I always ask the customer where he needs it for. You can not compare a DSO Nano scoop with a 20Mhz analog scoop! Then you don’t understand why they made this little opensource scoop! It has indeed it limitations! But if you just need to monitor a simple singal on the road and log it then I wish you good luck with your free 20kilogram 20Mhz scoop. No the DSO Nano is your best friend then! The little scoop is not for beginners who wanne learn. It’s a tool and you must use it like that. The DSO Nano is a realy nice scoop for a good price. It
But I also love the Rigol DS1052E Dave!
Did you know that it also can come with a logic analyser module!
Great to see someone make the point I have been making for years.
Some years back, when I was working at a TV Studio,Tek & HP would send their reps around to try to convince us to buy the(then) latest & greatest Digital Oscilloscope.
When we weren’t impressed,they would curl their lips,& explain to us “Technological
Dinosaurs” that we were used to seeing signals in one way,& that vector & dot displays were just as valid–blah,blah.
This is apparently the argument they had been programmed for ,& they didn’t realise we had real concerns about the performance of their devices.
We would then ask them to set the DSO to display two fields of a PAL video signal.
( This is a standard test to look for hum,poor clamping etc).
The display would look nothing like a video signal,with spikes sticking up everywhere,like the grass in my unmowed back yard.
The reason for this,is that when the time
base is set to display two TV fields,the total
time of the display is 40msecs.
Digital scopes commonly reduce the sample rate
a these speeds, as they cannot store enough samples at normal rate, with the result that any frequencies greater than the sample rate are not reproduced correctly.
A TV signal is a complex signal,with components up to 5MHz,hence the problem.
In many applications this would not be a problem,but in anything similar,it is a big problem.
Upmarket DSOs have overcome this to some extent,but the “cheapies” are still where HP & Tektronix were years ago.
Just wondering if anyone can advise me or has any knowledge of the Hantek DSO-1200 Handheld Oscilloscope?
Link For specs: http://www.hantek.com.cn/english/produce_list.asp?unid=81
I’m currently a Undergraduate student and in my first year studying Electronics and I’m after a reasonable entry level to intermediate scope that I can use at home but also something I can afford.
The Hantek is attractive because of it’s size, bandwidth (200Mhz) and the many functions it comes with. Also the price seems reasonable on ebay as they sell between
Mark,read my comment above,plus watch Dave’s original video.
The 300lb Gorilla in the room,which the people selling DSOs don’t mention,is that most of these devices reduce their sampling rate as the time/division is increased.
They have to do this,as their memory depth is inadequate to store the number of values that would be produced by sampling at the max
(advertised!)rate over say,20ms.
The better DSO’s try not to reduce the sampling rate too much, & hence are more useable.
The very good ones,which cost about as much as a small car,don’t have much of a problem at all!
If you are sure you will never have to look at a high frequency signal at a time/div setting much longer than its period,I’d say go ahead,but don’t be surprised if you get caught out somewhere along the line.
As in my previous comment,the main reason you may have to use a large time/div setting,is chasing AC hum on a HF signal.
There is a work-around of sorts:-
If the signal is riding on hum,there should be a thickening of the trace on the display when you look at it at its normal period.
If it isn’t noise at a higher frequency,which you can check easily enough,it is probably hum.
It still isn’t as intuitive as being able to look at it at a 50Hz/60Hz rate.
I can remember once where a RF amp I designed sometimes oscillated at 2GHz depending, and that caused a low-frequency ripple on the modulated output frequency. I just wish I had one of those reps there that were always trying to sell their DSO’s there, and ask him to prove to me that I can see it on a DSO. (We had lots of DSO’s, but not one could show that LF ripple.) I always have the fear that you can see effects like that on a DSO, but that I don’t know how to properly operate it. So yes, I agree, for some applications you can’t beat an analog scope, but for most digital work, I would grab my DSO, and yes, these low-cost USB-scopes are rubbish. The nice thing about some DSO’s is that you can sort-of use it as a low-cost audio-analyzer or spectrum analyzer in FFT mode.
Thanks for reminding us that sometimes newer isn’t always better. I found a local deal on a Hitachi Oscilloscpe V-1565 for $75 US. Any thoughts or history on this model?
Very useful contribution from vk6zgo re observing fast signals with slow timebase settings. I’m not familiar with the equipment; the problem described seems to be due to limited memory capacity, though maybe there’s sopmething else I haven’t seen. Does the issue of meaningless display in this case also apply to all PC-connected oscilloscopes; in principle massive storage, both RAM and disc, is available? Sampling for 50msec at 1GS/sec would generate 50 million data points if my arithemtic is right, not a great deal, but I don’t know what is involved in acquiring and storing data at this speed. As comment, the old PC-connected oscilloscope I have, a Tektronix TWD120, has I think good attenuators and triggering, but doesn’t use the PC’s storage and is limited to a maximum of 5,000 data points.
A website about (not by) Jim Williams of Linear technology gives some other points where analog are said to be better than digital scopes. Without adding any comment of my own, here is the list (detailed discussion linked from URl below):
Trace clarity (resolution and spot size)
Sensitivity and bandwidth (and noise floor)
Overdrive resilience (of sampling plug-ins)
Repairability and inspiration
When the advantages of analog scopes are discussed here, is “analog scopes” implied as the completely analog ones or rather as some old scopes in opposition to the contemporary DSO’s?
For example, can Tektronix 2235 be considered as analog scope?
I’ve watched EEVblog #13 (Digital storage Oscilloscope Tutorial) and have the answer on my question now. Thanks.
Leave a reply