Checking audio amplifier distortion with oscilloscope spectrum analyzer
HTML-код
- Опубликовано: 11 июл 2017
- This video explains how I setup my scope to check the distortion of the amplifiers in my review videos. This is the 2nd part of my test signal video...
• DIY audio amplifier te...
Here is a link to more in depth FFT explanation:
www.edn.com/design/test-and-me... 
Наука
How can anyone not like this video and the information presented therein? John is a great 'non-BS' audio connoisseur, who sticks to the facts and figures, unlike many others who just don't know what they are talking about.
You keep stuff simple thats great ... I relise your simplifing the stuff for others,the more i watched so i relise now what your actually up too here wich is a very awsome positive thing because your helping ..this makes me very happy to see ...and your doung a great great job...great teacher ..keep it up jhon...73 from justin Oklahoma DX1 ..
First !
Sorry for the lack of videos lately. I do have an amp board to test, another on the way, a couple chip amp test videos, teardowns and more. Hang in there guys!
No worries, take your time :-)
thanks John, I used your method to measure the noise in a project, I will document it a little whenever I finish it. Anyway could you please explain in a little more detail the calculations needed for the power specs of an amp, you always make that fast. My project involves a couple of LM3886, a OPA2134 for preamp, an MCP4261 as attenuator and I would use an Arduino to listen to a remote control, switch between several relays and turn on/off a raspberry pi (with a Wolfson DAC) as one of the sound sources. I have really sensitive speakers, so I'm not concerned too much in power, moreso in the possibility of noise in all this mess. I will not use any switching psu, just old plain trafos a linears, and when everything is off, I'll use a really small trafo for the standby mode, which is just the Arduino. I really enjoy your videos.
JohnAudioTech aaaawwwšqwwwwaw
I love your Videos! Not although but because you dont have a fancy pancy High End Lab. Your content is easy to follow and you are explaining stuff in a way, a lot of people can understand. I love it. Keep it up :-)
John: Great video. Saw part 1 as well. I really like that pilot as a reference. I am going to try it. Cheers.
This is a good quick seat of the pants method, for example in a repair to verify a level that is too ambiguous by ear but still indicates a problem. I work on a lot of guitar amps and most people working on old tube gear use very little in equipment but judge by playing. Few use a spectrum analyzer or any repeatable signal, more by feel and experience. I work on so many different types that I like to log reading in a database as a reference state in case it comes in again. Most gear does fail periodically when used in touring or stage use, amps, power amps cable lead hard life on the road. As the end of a tour a company might bring 10 or 15 amps for me to check and comparing past performance to the current readings tell me if there is a possible change or problem.
For home hobbyists, a decent digital multimeter which are very reasonable now, $25-50 for one that reads AC volts in True RMS which is best for audio use.
An Oscilloscope. The new digital scopes are a real powerhouse of features and have a lot of built-in measurements that can reduce the number of test instruments needed by 2/3rds or more. I am a test gear fanatic so have done of precision instruments but they are not essential.. In my old shop just doing pro audio, I had 23 techs and 2 full benches of test instruments for each tech. Over 500 instruments because I collected them all and refurbished them as industrial surplus. Now, a modern Chinese-made $300 scope can provide an amazing about of data. My bargain digital scope is hard to bear, a Rigol 1054z that has 4 channels and a large number of advanced features you can add simply by typing in a code, published on many sites and RUclips to turn on many features they charge more for at the time of purchase. Definitely check that model.
Used older analog scopes represent low-level noise better than digital scopes and are good value. Any model like a Tektronix 465 or 2235 or 24xx series scopes are wonderfully built and outlast you and are reasonably priced as ham radio swap meets or eBay but better to find lone locally. They were expensive in the day, , all good test gear was. My 1975 Motorola RF Service monitor was $28,000 when I bought it new and my house at the time was $17000 on 10 acres in California! My Sound Technology 1510a Digital analyzer for analog tape decks in 1985 was $14,000. I used that to calibrate the large 24 track Studer and Ampex tape decks every morning in my 3 studio complex recording studio I had for 20 years.
A good noninductive dummy load, usually 4, 8 and 16 ohm rated at higher than the maximum power of any of the amps you will work on. These can be made buy buying a few high power non-inductive resistors, You can get by with cheaper standard wire wound resistors, 100 ro 250 watts are good choices
A good sound source which is a clean low distortion sinewave generator with a wide frequency and level range. I have some very nice expensive ones but for everyday use it is had to beat the old HeathKit IG-5218. It has all the features one would want, the accurate level meter in db and volts, accurate decade switch frequency selection, low distortion, with a wide range of accurate attenuation of output. So when selecting a level, and frequency you do not have measure them, the controls and meter are accurate to 1% or better with low 0.013% harmonic distortion. There are modes on the internet for modifying them for even lower distortion by making an opamp simple meter isolation amp that removes the distortion caused by a bridge rectifier across the before the output attenuated, I have one that is 0.0068% at 1khz and +10 dbm out, It is not likely you would never need or hear that difference.
A constant temperature soldering iron.
Those are enough to handle most repair or adjustment tasks with audio.
You can also get digital signal generators but they usually are not as versatile as the old Heathkit IG-5218.
If you have a good sound card on your computer you can use it along with a free program for a very accurate signal generator, meter, counter, and Spectrum Analyzer...the only problem is that you are limited to the output level of your sound card. If you have an outbard high resolution sound card interface used on balanced line inputs and outputs you can handle wide levels and usually better specs
I have a $4800 spectrum analyzer and this program does better and free www.sillanumsoft.org/download.htm It will do frequency sweeps to plot frequency response by distortion, and two channels to compare two channels in real-time.
Two more things, a regulated two-channel DC Voltage and current metered power supply This will be used on your breadboard amp and electronics projects until the circuit work then you use batteries or build a dedicated power supply for it.
And if working in AC powered united a Variac, which is a variable power transformer to you can dial in the ac mains voltage you want. This is important when checking an amplifier that might be damaged, you can monitor the current it pulls as you increase the mains voltage from zero.(only for units that have power transformers and not Switchmode power supplied some modern amps use to make amps a lot lighter. Your computer has a 300-400 watch switch mode power supply and you could not raise the voltage from zero, Just turn it one since it has its own overcurrent protection circuit.
thanks for sharing your experience and all those amazing tips.
The key to increase resolution is "oversampling and decimation".
Also, any speaker does give out distortion well over 1%.
Therefore, there is no use for 0.01% distortion in audio.
Bandwidth limitation is much more important for the human hear.
I have to say I have an opposite take on the equipment setup I have all the gear and no idea how to use it. You do amazing work with modest test and equipment your video are a pleasure to watch. Thanks for sharing. I have Been testing a low cost amplifier on my channel I know nothing about audio. If you get a minute drop in and say hi. All the best to you and the family for a happy Christmas regards Chris
I have the same scope. Thanks for the tips.
John, have you ever talked about ground referencing between the scope, the amp, and the source? I think I blew a couple of fuses on an old receiver monkeying about like this with dummy loads. I was using my pc audio out as a source. I assume it was a grounding issue, but I think I had all the right polarity on my leads. Hmmm. Appreciate the channel very much, keep up the great work.
Thanks for this great video! Do you have any example waveforms that’s show some distortion?
I have now looked at many videos about this subject, but this is the first time i really get it, so many thanks:-)
The term u want to use is bandwidth..and resolution.... And rhe scope measures voltage over time..the anlyzer measure amplitude over FREQUENCY.... Hryz per dv... Dbm per dv... And linear log or non linear ligrimithic... ...
Funny, I was just taking some distortion measurements last night. It seems that modern audio sources can be the source of noise too, my laptop and phone all have lots of high frequency noise on the audio output and this shows up as a fuzzy line riding the waveform. My tablet on the other hand had a relatively clean output.
Next time you test an amp can you have a look at the rf content in the signal as for some reason my DS1052E always shows a rising edge at the end of the spectrum analyzers blue line, I don't have a true signal generator so can't be sure what's causing it.
Hi John i was thinking about upgrading to the Rigol 1054z as at present im using an old Crt Kenwood CS-4125A. As far as i can tell the FFT on the 1054 is aweful much worse than 1054 which i thought looked pretty good. Have you any thoughts on this or have you looked into a new digi scope. Been watching your vids for the past 3 months and have learnt a fair bit. I've been power testing many of my Tube guitar amps some of which i've built myself.
Kind Regards
Matt
John, thanks a lot for this upload. This was very useful info.
can i measure output of amplifier connected on dummy load with scope if i have 100v+- on rail .is it safe or is this gonna fry my scope?i have 1000w amplifier and i wanna see where is the distortion limit?
The Hantek scopes are affordable. I was able to get a 200 MHz with function generator for $225 shipped. It doesn't matter if you buy a 60 or 100 MHz Hantek can easily be upgraded to a 200 MHz using a urat takes less than 30 minutes. Tones of info on how to do it online and on RUclips. Great video 👍.
Great video!
great stuff, thanks for the info. John, could you elaborate on 1/256 (0.0039 the logic underneath) thanks again, great easy tool assess audio toys :)
Over sampling play some role in this demo, but more importantly, the FFT display the amplitude in power, so 8-bit vertical resolution in voltage gives 48dB of dynamic range so seeing distortion in 0.2% level is not that hard. In this case you are probably limited by scope front end noise rather than the dynamic range of the FFT.
can you explain how you calculate the percentage with FFT ?
Thanks for your work.
i think he's playing files that already contain the said distortions..
Great video John. I am very interested to test this out on my scope. My I ask where did you get the Audio test tones used in the video?
I made the tones using a free software called Audacity. I have the tones on my Google drive for downloading. This video explains the tones and has a link to the files.
ruclips.net/video/cR1Tz8dQ1NE/видео.html
JohnAudioTech wow. Thanks John. I'll let you know how I go. Check out my bench amplifier on my channel if you like
Hey, I'm not sure if you could help me out or not. I have a Fluke Scopemeter 105b II. I use it solely for the purpose of looking at analog audio signals. When using it to test a low voltage signal coming from the HU from 0VAC to 5VAC the Scopemeter displays the signal perfectly. When I use it to look at the signal coming from an amplifier 10VAC to 50VAC the signal looks crazy. Do I have to change something on the oscilloscope to look at higher voltage analog sine waves with out the waveform looking like it's having an aneurysm?
I'm not familiar with that scope but you have to set the time base and amplitude for the signal you are trying to view.
JohnAudioTech on my scope, you can hit an auto button which calculates in displays the waveform, but even after adjusting the time and amplitude it still displays incorrectly and sometimes even displays an exact wave further ahead of it, it's weird.
Thanks, good video John. How have you set up your test signal with a percentage marker signal? I’d be interested to set something up like this for testing my audio amps. God bless, Bill.
I used Audacity to make the signal. The marker is made to whatever the percentage amplitude is needed but you must allow headroom with the main signal so there is no clipping when they are mixed.
I need to try this! Can you tell me what the "pilot tone" is all about? Is that a tone that you're injecting for a reference? How is it created? .. Great video! Thanks for posting! :-)
Thank you for your amazing videos! I've learned so much! Newbie question: Are you just using a 1K tone from the audio player or are you using a special tone that includes the pilot tone with distortion? Thank you!
The pilot tone is included in the signal at 1% of the 1KHz tone and at a non-harmonically related frequency.
@@JohnAudioTech Thank you! Is this something I need to create or is there a resource to download the audio file?
Great lesson.. But if I apply this to my amplifier 2000W..would it be frying my scope? Very little knowledge testing or checking amplifier with scope
What is the peak input rating of your scope? If the amp is honestly rated, 2000w into an 8 ohm load would be 179 volts peak (or a bit more if there is headroom in the rating). 2000w into a lower impedance would be a lower peak voltage. The two scopes I have are rated 400 volts peak.
What exactly are harmonics? i`ve heard people talking about them but i never really understood what they actually are.
This question would be a good subject for a video. It would take a long explanation to cover it here.
I would really appreciate it, since I really enjoy watching your videos cause I just like the way you explain things in an easy-to-understand-way. Not to mention that I like playing around with electronics and audio a lot and did it since I was a child.
Thanks for the answer!
Why didn't you use dBVrms vertical scale?
It compresses the view on the screen too much.
Only high amplitude signals, right? That's the main reason to use log scale, isn't it? I really couldn't see the advantages of using linear scale to do that. Are you sure about that ? Could you explain your reasons a little further? Thank you.
omg my scope is in this video, give back my scope hehe. He is one of many. It's like audio equipment, you never have too much :)
Have you tested amplifier power output and distortion with lower frequency tones? For example at 50Hz. I'm interested how TDA7297 performs at lower range.
Solid state amps usually have no issues with lower frequencies. It is the higher frequencies that are the problem. Even so, most well designed amps distortion is low there too.
Love your videos John 1 thing i wanted to ask you-did you "overclock" or "hack" your oscilloscope to 100Mhz yet? Coz you know it can be changed to that another model 100Mhz one.Check DS1052E hack on google!
TIL what “long in the tooth” meant.
thanks for that explanation!
So this is a poor man's way of measuring distortion instead of spending $1000's on fancy distortion analyzers. Love the simplicity and common sense.
I only have a very basic analog scope.