Buying my first scope, advice sought

[hiwatt25]

Hello all,

This is my first post, having found the eevblog site just last night while doing research for my first scope purchase.  I'm hoping some of you with more experience than I might be kind enough to share some feedback with me based on my scope needs.  

In a nutshell, I build and fix guitar effects pedals and would like a benchtop oscilloscope for use in troubleshooting circuits.  I do a little bit of circuit design but it's very limited as I'm still just a beginner.  I'm much better at designing the PCBs than the circuits.  I'd like a scope that will foot the bill for audio related electronics.  

Given my budget and the light duty nature of what I do, I'm primarily interested in two scopes.

The first is the Owon EDU5022
http://www.tequipment.net/OWONEDU5022.html

And the second is the Rigol DS1052E
http://www.tequipment.net/RigolDS1052E.html

My questions are these, will either of these scopes serve me well with regard to audio testing/troubleshooting and if so, should I save the $100 by buying the Owon and then use that hundred bucks to buy a nicer multimeter than the one I have now or am I better served by the Rigol?  

I must admit I love hacking stuff and the possibility of hacking the Rigol to make it 100mhz (while I may never need that functionality) excites the hell out of me.  I started building effects pedals because I wasn't satisfied knowing that something worked, I needed to know why.  Once I started down that road, I was hooked.  

Anyhow, any input is greatly appreciated.  I think I may be here quite a bit.

Cheers!

[allanw]

I believe even a cheapie 20MHz analog scope for sub-$100 will satisfy your needs for audio circuits. The scopes you're considering are more than adequate.

[alm]

Yes, a decent 20MHz scope would probably suffice (unless you get into digital audio stuff), but that doesn't mean any digital scope will do, analog scopes might actually be superior for some audio work. An issue with audio is low level signals, so input noise is an issue (some digital scopes have tons of noise at low vertical ranges). Good analog scopes with a fine trace (some of the older high quality low-bandwidth ones) have much more resolution than a QVGA display with 8-bit ADC.

I have no direct experience with either, but what I've read about the Owon scopes (I think it was in Elektor) wasn't very positive, so I'd prefer the Rigol over the Owon.

[EEVblog]

I third the 20MHz analog scope idea. It will be all you need, and will almost certainly provide superior performance and display for low level analog use.

If you do get a digital, get the Rigol over the Owon.

Dave.

[hiwatt25]

Thanks for the input guys, as it turns out my dear friend Ben Verellen at Verellen amplifiers just GAVE ME a Tektronix 2225 50Mhz scope!  It looks just like the one in this auction.  What do you think, should it do the trick?  If so, Ben just saved me like four hundred bucks and will be receiving some nice Pacific Northwest microbrew!

http://cgi.ebay.com/Tektronix-2225-Oscilloscope-50MHz-2-Channel-/300438015842?cmd=ViewItem&pt=BI_Oscilloscopes&hash=item45f3804f62

He's not sure if it needs to be calibrated or not, are there people that can calibrate scopes online?  I wouldn't even know where to look for that kind of service.

Thanks again, everyone~

[ThunderSqueak]

Should work great  Congrats   That is a perfect scope for that type of work, beat the hell out of what I started with !

[hiwatt25]

I'll have to buy my own probes for the 2225, does anyone know what probes to buy?  I was looking at tequipment.net because I think I'm going to buy a new multimeter as well. 

Thanks again!

[EEVblog]

Thanks for the input guys, as it turns out my dear friend Ben Verellen at Verellen amplifiers just GAVE ME a Tektronix 2225 50Mhz scope!  It looks just like the one in this auction.  What do you think, should it do the trick?  If so, Ben just saved me like four hundred bucks and will be receiving some nice Pacific Northwest microbrew!

http://cgi.ebay.com/Tektronix-2225-Oscilloscope-50MHz-2-Channel-/300438015842?cmd=ViewItem&pt=BI_Oscilloscopes&hash=item45f3804f62

He's not sure if it needs to be calibrated or not, are there people that can calibrate scopes online?  I wouldn't even know where to look for that kind of service.

The 2225 is in fact almost the PERFECT scope for analog work. It has a specific low noise analog front end that goes down to 500uV/div.

I've been meaning to pick one up myself for just such low level work.

You can check the cal yourself by comparing against a multimeter.

Dave.

[alm]

You can do a basic performance verification (except transient response) with a function generator, adjustable power supply and DMM (check vertical and horizontal accuracy/linearity). If the probe cal signal looks good after adjusting the probe (nice square wave with correct amplitude and frequency) and you're able to get a bright trace with good focus all over the screen, you're probably fine. A scope (especially an analog one) isn't a precision instrument, so calibration isn't that important in my opinion. Most cal labs can do scopes, although they probably won't be happy with an analog one with lots of trimmers. Plugging in the GPIB/USB/LAN cable and press 'start' (i.e. most digital scopes) is much less work. Not sure how online calibration would work, you'll have to get your scope to the cal lab somehow (unless you pay extra for calibration at your location).

About probes, Choosing a probe. For audio, you'll probably also want 1x probes. I'm not a big fan of the switchable type, but for 50MHz (and especially audio), they're probably fine. I prefer real Tek probes that I buy used on Ebay, but again, for 50MHz, it's probably not such a big deal.

I've been meaning to pick one up myself for just such low level work.

Another alternative for low-level work is the Tektronix AM502, which is basically a 'stand-alone' version of the 7A22. It's a differential amplifier (also works in single-ended mode) with a gain up to 100,000 (10uV/div with the scope at 1V/div), an excellent CMRR, adjustable DC offset and low/high pass filters. The only downside is that the bandwidth is just 1MHz, but that's plenty for audio. You can feed the output to any scope. If you want two channels, you would need two of them. It does need an TM5xx/TM5xxx mainframe, but it's smaller than yet another scope.

[DJPhil]

Welcome, and congratulations on the scope!  It's perfect for audio work, I'm jealous.

I'll have to buy my own probes for the 2225, does anyone know what probes to buy?

In truth, just about anything out there will do for audio work, even the ultra-cheap stuff. Without digging to deep at tequipment, something like this should be fine. Two will get you started and allow you to use both channels as well as experimenting with a separate trigger input.

Further reading:
Tektronix has two documents you might find useful, 'The ABCs of Probes' and 'The XYZs of Scopes'.
The Boat Anchor Manual Archive (BAMA) mirror turned up a user manual.

Hope that helps.

[hiwatt25]

Wow!  Thanks for all the great help.  I"m excited to get started.  There's a fella here in town, on craigslist, selling a function generator that I think I'll pick up tomorrow...especially since I won't be buying a new scope now.  Woooo.

Thanks!

[migsantiago]

Top 10 things to consider when selecting an oscilloscope.pdf

http://www.4shared.com/document/OkjZEv-s/Top_10_things_to_consider_when.html

[chscholz]

I believe those are very good point. As always it depends what you do

8 bit ADCs are not good enough for some audio applications. I'd also like to caution that a "thin trace" does not necessarily mean low noise. You really need to look at noise spec in the data sheets. However, manufacturer's specs are notoriously hard to compare since they are often only valid for very short acquisitions and often deteriorate rapidly for longer acquisitions. As always read the fine-print and various footnotes in the spec sheets.

Many scopes offer "enhanced resolution" (aka. eres). The screenshot below show how much e-res (set to 3 bit) improve measurement results. The value measured it the peak-to-peak value of the time interval error.

Yes, a decent 20MHz scope would probably suffice (unless you get into digital audio stuff), but that doesn't mean any digital scope will do, analog scopes might actually be superior for some audio work. An issue with audio is low level signals, so input noise is an issue (some digital scopes have tons of noise at low vertical ranges). Good analog scopes with a fine trace (some of the older high quality low-bandwidth ones) have much more resolution than a QVGA display with 8-bit ADC.

I have no direct experience with either, but what I've read about the Owon scopes (I think it was in Elektor) wasn't very positive, so I'd prefer the Rigol over the Owon.

[Ronnie]

Yes, a decent 20MHz scope would probably suffice (unless you get into digital audio stuff), but that doesn't mean any digital scope will do, analog scopes might actually be superior for some audio work. An issue with audio is low level signals, so input noise is an issue (some digital scopes have tons of noise at low vertical ranges). Good analog scopes with a fine trace (some of the older high quality low-bandwidth ones) have much more resolution than a QVGA display with 8-bit ADC.

Also I'm in search for my 1st personal scope and will be using it in the design of discrete class D audio power amp just like the one designed by Philips here http://www.nxp.com/documents/user_manual/UM10155.pdf. With PWM frequency specification of 400 kHz may I know what scope bandwidth is needed for this kind of audio work? The output before the LC filter contains harmonics of the PWM frequency which is way above the frequency response of a typical class AB power amp.
For microphone preamp design is it really analog scope the only option? Are there low priced digital scopes capable for audio work?

[alm]

With digital(ish) signals, transition (rise/fall) time is usually the important parameter, not repetition rate, I couldn't find rise time specs from a quick glance at the document. My guess would be that 20MHz is enough, but more bandwidth rarely hurts. It's hard to find a decent scope with less bandwidth than 20MHz (unless it's really old), so that shouldn't be an issue.

I believe a microphone signal (before the preamp) is pretty low level, so a cheap digital scope with lots of noise at high sensitivities is not a great match. If I remember correctly, Dave compared the noise of the Rigol scope with a Tek TDS-2xx, which is known for pretty bad noise at low levels, and it was fairly similar. But a cheap digital scope is infinitely better than no scope. Modern digital scopes can do most things that analog scopes could, but the cheap scopes don't usually have all these features to emulate analog scopes like intensity grading and lots of acquisitions per second. I think that a $100-$400 analog scope would be a better choice for audio work than a $400 digital scope.

[EEVblog]

I believe a microphone signal (before the preamp) is pretty low level, so a cheap digital scope with lots of noise at high sensitivities is not a great match. If I remember correctly, Dave compared the noise of the Rigol scope with a Tek TDS-2xx, which is known for pretty bad noise at low levels, and it was fairly similar. But a cheap digital scope is infinitely better than no scope. Modern digital scopes can do most things that analog scopes could, but the cheap scopes don't usually have all these features to emulate analog scopes like intensity grading and lots of acquisitions per second. I think that a $100-$400 analog scope would be a better choice for audio work than a $400 digital scope.

I agree.
If I was working on audio gear I'd want an analog scope on my bench, no contest.
I don't know what all the fuss is about really. If you want a $400 digital scope, get one. A 2nd hand analog scope is only ever going to set you back $50 or so. You can have both!

Dave.

[DJPhil]

With PWM frequency specification of 400 kHz may I know what scope bandwidth is needed for this kind of audio work?

Check out Wikipedia's page on the square wave. Assume you want to cover as high as the eleventh harmonic of 400kHz, you'd need to be able to see a 4.4MHz sine. Now this is a simplified (some would say alarmingly) approximation to be sure, but it helps show that for what you want to do a 20MHz or even a 10MHz scope is likely to be adequate for basic testing. In a way you're doing one of the more forgiving tasks where harmonics are a concern, so in my opinion you've got a bit of wiggle room. That said, more bandwidth isn't a bad thing.

For microphone preamp design is it really analog scope the only option? Are there low priced digital scopes capable for audio work?

I wouldn't say it's the only option, but it's a good one. I tend to side with Dave on digital scopes and say that if you're going to buy a new digital scope you should consider the Rigol (or the recent Instek competitor) and skip the cheaper hand held stuff. Picking up a second hand analog at half that price or less is ideal for audio work. You may run into low level signal and noise limitations with a cheap digital scope, especially a $100US USB model. There are strengths to both analog and digital and for audio work analog is preferable, but if you know the limitations of your hardware either one can work swimmingly.

My opinion in short, a second hand analog scope in good repair will blow a $200US USB scope out of the water for audio work, especially class D work. Even up near $400US when you start considering the Rigol and friends you'd have to think about the features you're gaining versus the difference in resolution and sensitivity.

Hope that helps.

Edit: I type slow, the pros got there first. Makes me feel better that I evolved the same opinion they've expressed.

[chscholz]

You need to be careful with the 500 uV/div specs though.

First of all, the 0.5 mV/div setting limits the bandwidth to 5 MHz. (page 4 item 14 of the operator's manual 070-6298-01).
Second, table 5-1 (Electrical Characteristics, p. 18) states under "deflection factor range": "5 mV per division to 5 V per division [...]. Sensitivity increases to 500 uV per division with X10 vertical magnification". In other words, for 0.5 mV/div you get about the same noise performance as in the 5 mV/div setting but of course the trace is 10x larger. This can be a useful feature in some cases.


[/quote]

The 2225 is in fact almost the PERFECT scope for analog work. It has a specific low noise analog front end that goes down to 500uV/div.

I've been meaning to pick one up myself for just such low level work.

You can check the cal yourself by comparing against a multimeter.

Dave.


[/quote]

[chscholz]

For analog scopes scope risetime is pretty close to 0.35/(Scope bandwidth), for digital storage scopes tau_r is anywhere between 0.35 to 0.5 /bandwidth.

For digital audio digital storage scopes you can also decode the audio signal, reconstruct the analog waveform, perform your A-weighted filter, save and play the analog waveform, trigger on specific addresses, etc (see below). In addition to rise/fall times you should make sure to get large enough memory depth.

With digital(ish) signals, transition (rise/fall) time is usually the important parameter, not repetition rate, I couldn't find rise time specs from a quick glance at the document. My guess would be that 20MHz is enough, but more bandwidth rarely hurts. It's hard to find a decent scope with less bandwidth than 20MHz (unless it's really old), so that shouldn't be an issue.

[...]

[Jon_Paul_Clarke]

Evening all,
                I've been trying to look for other models of CRO that have the golden "500uV/div" sensitivity for the channel inputs, but I have only gleaned a few model numbers from Dave's video rants/CRO p0rn/Interweb debates. They are:

1. The venerable Tektronix 2225
2. The Hybrid beasty Tektronix 2211
3. Possibly the 2214?
4. The sexy Tektronix 2440
5. The Tektronix 7000 series (with appropriate plug-ins)
6. The rock solid Hitachi V-1050F
7. Others?...

This is about as many as I could get, does anyone else no of other Models/Makes that have this input sensitivity?
Also, I'm trying to get my hands on a 100MHz model for all-round work.

Thanks in advance
Jon