I have dabbled in audio over several decades and am looking for the best oscilloscope for audio measurement and diagnostics for <$300. For this purpose I understand that memory bit depth is more important than high bandwidth. For example a 50MHz or especially a 100MHz bandwidth would be more than adequate, while the Hantek's 48Kpts (24 per channel) may be inadequate for useful fft, especially when trying to measure harmonic distortion. There is a new Rigol DS 1202z for $299 that has 24Mpts of memory.
So my questions are,
1) Are my assumptions correct for the uses I described re: bandwidth and memory?
2) What is the minimum bandwidth and memory I should consider for the uses I described?
3) What other specs should I look for?
4) Can you suggest any specific brand/model that would be particularly well suited for this purpose? I do not want to hook up a box to my computer. My computer stays in the house and my "lab" is in my garage and I can't spend another $1k on another computer.
I also just learned that a scope with a high number or refresh or retrieval points is better with FFT. Most lower priced scopes have ~1,000 points. One scope was pointed out on the EEV blog, the Instek GDS-1054B (~$300) that has 1M points and 50,000 wfms/s while others at this price point have 30,000 or considerably less.
I am unfamiliar with these factors, so in noob terms, I assume it has to do with refresh rate and the number of pixels it can display in a unit of time. FFT on a scope requires lots.
Typically probes that can switch between 1X and 10X come with low end scopes while the scope's probe matching capabilities are wider than that.
1) When you recommend "100x or 1000x magnification probe support", are you suggesting using 100X or 1,000X probes?
2) As I understand, the "x" really means attenuation or division by x times. So, an amplifier's line level input into a scope should be attenuated (divided) 100 or 1,000 fold?
3) For example, I would be testing an amp that generates 120 watts RMS per channel, with a very low distortion spec. I'm guessing a mere 10X probe would overdrive the scope but dividing that power by 100 or 1,000 would bring the scope input into a more manageable range - is that the point?
4) Is it a matter of not blowing up the scope or just achieving more meaningful results?
Both the Rigol and the Instek I mentioned have that probe option capability, something like a range of .01X to 1,000X.
Scope uses a resistor in series to get the signal in a small amount and that can be measured in higher voltage resolution. For example, If you want to see a 5V signal that has a ripple of 10mV, you couldn't see it in a 1x probe. Why? Because of the range. Suppose you want to see 50mV per division and use 10x magnification, then you could see it in 5000mV / 10 = 500mV per division range. Further 100x will allow you to get the value in 50mV per division range.
For audio testing, it is important, because, often audio noises such as 50 or 60 cycle hum is a very low noise and to see it carefully you required higher magnification.
1. Yes I am suggesting that.
2. Exactly.
3. Ummmm not exactly. For RMS you need to calculate it by the RMS function of the scope. However, A scope has a maximum input rating of 250V or 300V or whatever the spec you get, do not override it at any situation.
4. Getting an observable and meaningful result.
I'm not understanding the 3. answer. Let me break it down.
1. The purpose of a 100x or 1,000x probe is to reduce the voltage of the amp's speaker level output to a range that results in a more accurate scope measurement than what a 10x probe would provide, right?
2. I'm not sure what you mean by "you need to calculate it by the RMS function of the scope."
a. Calculate what? What is "it?"
b. What is the "RMS function of the scope?"
c. Could you provide an example of scope and amp specs that would require various "x" probes, e.g. 100x, 1,000x or one model on Amazon that is switchable between 50x and 500x.
1. Yes.
2. 'It' RMS
3. If you want to calculate RMS of an signal you can manually calculate the rms output by mathamyetically using pen and paper, that is obvious. But maximum scopes has a math function, once trigger it will show the RMS of the signal. Modern scopes (digital) has this.
4. The x10, x100, x1000 is available as a selection switch in the probes, also, you after setting the switch for example x100, you also need tp configure it in oscope probe settings. You can buy a scope with maximum x1000 probe input and could separately purchase the x1000 enable probe.
1. Yes.
2. 'It' RMS
3. If you want to calculate RMS of an signal you can manually calculate the rms output by mathamyetically using pen and paper, that is obvious. But maximum scopes has a math function, once trigger it will show the RMS of the signal. Modern scopes (digital) has this.
4. The x10, x100, x1000 is available as a selection switch in the probes, also, after setting the switch for example x100, you also need to configure it in oscope probe settings. You can buy a scope with maximum x1000 probe input and could separately purchase the x1000 enable probe.
Sourav Gupta
Permalink1. You are right. Choose something that has more than 24Mpts. Since the audio signal does not cross 20 Khz bandwidth, 20-50 Mhz is perfect. (You will get additional free bandwidth from 20 kHz to 20-50 Mhz for transient analysis).
2. Memory is required due to the complex agronomy of the signal. If you test the signal integrity test or THD, you need adequate memory in the scope to support the ADC response time of the scope. I will not bothered about the bandwidth.
3. Choose something that supports 100x or 1000x magnification probe support.
4. Nothing in my mind. Look for Hantek. You can also look for old Tektronix.
- Log in or register to post comments
Joined February 12, 2018 696Monday at 02:11 PM