Which oscilloscope probes for testing audio amp speaker outputs

[gfmucci]

I have an integrated audio amp that produces 160 watts into 4 ohms per channel that I would like to test for "clipping point" with an oscilloscope.

I was told that when connecting and testing speaker outputs from an integrated power amp that 100x or 1000x probes should be used.

Assuming no use of an inline resistor from the amp speaker-level outputs, the output to the scope would be ~126volts.

How can I know what value probe to get?  I see three types available on Amazon:  100x, 1000x, and a switchable between 50x and 500x.

Alternatively, I've seen high power inline 4 or 8 ohm resistors available that are used for this type of test.  Could 10x probes be used if this type of resistor is used?  With a 4 ohm inline resistor the speaker output would be around 25 volts.  Which is the best approach, the alternate probe approach or the inline resistor approach, and why?

I understand that since the amp has a 3-prong grounded power plug, I would need either an isolation transformer for the amp to plug into, or isolation probes to plug into the scope. The isolation transformer is much cheaper than isoloated probes.  Which is the best approach and why?

[Benta]

A standard x1/x10 probe is fully sufficient for this job. If the amp has a bridged output, you might need two probes. Amp earth and 'scope earth should be the same.

"I was told" makes my hair stand up.

[TimFox]

Be sure to check:  most amplifiers of that specification have "bridged" output, where neither connects to ground and it will be seriously unhappy if you ground one.
Also, 160 W into 4 ohms is 72 V pk-pk, or 25 V rms. or 36 V pk-pk on each side of a bridged output.  As Benta also stated, a x10 probe will get that down to 7.2 V (0r 3.6 V) pk-pk, which is quite reasonable for an oscilloscope input.

[gfmucci]

Be sure to check:  most amplifiers of that specification have "bridged" output, where neither connects to ground and it will be seriously unhappy if you ground one.
Also, 160 W into 4 ohms is 72 V pk-pk, or 25 V rms. or 36 V pk-pk on each side of a bridged output.  As Benta also stated, a x10 probe will get that down to 7.2 V (0r 3.6 V) pk-pk, which is quite reasonable for an oscilloscope input.

So you guys would recommend that I use the in-line high powered (e.g. >100 watt resistor) to reduce the voltage down to ~25, and using the 10x probe as opposed to higher attenuated probes without inline resistance?

How about using the isolation transformer to plug the amp into as well to be safe?

[Benta]

Be sure to check:  most amplifiers of that specification have "bridged" output, where neither connects to ground and it will be seriously unhappy if you ground one.
Also, 160 W into 4 ohms is 72 V pk-pk, or 25 V rms. or 36 V pk-pk on each side of a bridged output.  As Benta also stated, a x10 probe will get that down to 7.2 V (0r 3.6 V) pk-pk, which is quite reasonable for an oscilloscope input.

So you guys would recommend that I use the in-line high powered (e.g. >100 watt resistor) to reduce the voltage down to ~25, and using the 10x probe as opposed to higher attenuated probes without inline resistance?

How about using the isolation transformer to plug the amp into as well to be safe?

You're overcomplicating this totally.

If you want to test the amp under load, then use your 4-ohm power resistor as if it was the loudspeaker. I've no idea what you mean by "in-line".

The output of your amp will be isolated from the mains by design (if it's not, commit it to the landfill immediately and sue the company that made it for x million dollars).

Connect the probe point to one side of your resistor and the probe ground clip to the other side.

Do the measurement.

All the rest is "Weekly World News".

[gfmucci]

"I was told" makes my hair stand up.

That's why I'm asking here.  But then again, when I am told "X" here, I will have to still say "I was told."  So I guess hairs will be in a perpetual state of standing up.

[Benta]

"I was told" makes my hair stand up.

That's why I'm asking here.  But then again, when I am told "X" here, I will have to still say "I was told."  So I guess hairs will be in a perpetual state of standing up.

Fair point, my apologies. Just always start with common sense and KISS, then everything evolves nicely.

[Gyro]

'Integrated amplifier' sounds consumer. If so, be aware that it probably won't stand running to the clipping point for any significant period of time, especially if you're driving both channels.

[bdunham7]

There are some points that aren't clear here.

First, why do you think your amp will put out 126 volts?  That seems unlikely.  Make/model?

Second, there's absolutely no point in testing an amplifier without a load and I'm not sure what you mean by 'inline resistor'.  Typically you use a load resistor similar to the speaker impedance (8 or 4 ohms typically) in place of the speaker.  You could just use a speaker but it would get very loud and their impedance is more complex and variable.  A good first project would be to make a 4/8 ohm load bank.  You need 8 50W 8R power resistors from All Electronics (a couple of bucks each) and  a chunk of aluminum heat sink.

Third, probes are just not an issue.  The standard 1X/10X probes you get with any cheap scope will suffice and in the vast majority of cases, no probes at all would work just fine. The scope has an input impedance of 1M (a million ohms) and many will read +/-40 volts p (80 Vp-p) directly--enough to likely spot your clipping. 

As for grounding, don't rely on an isolation transformer unless you clearly know what you are doing.  An isolation transformer with 3 pronged plugs will still pass thru the ground, which yields you exactly nothing in this case except maybe some blown fuses.  What you need to do is verify that you have a common ground (aka not bridged) amplifier where the negative speaker terminals are tied to ground.  You very likely do have this type of amp, bridged isn't that common.  If not, you have additional complex issues before you scope anything.  In case it hasn't been mentioned yet, PLEASE watch Dave Jones' video "How not to blow up your oscilloscope" until you understand it thoroughly.

[bdunham7]

Connect the probe point to one side of your resistor and the probe ground clip to the other side.

Bad advice.  On many common amps, there's a 50% probability you just fried the outputs, blew a fuse or melted your probe ground.  My hairs are standing up.

[gfmucci]

Be sure to check:  most amplifiers of that specification have "bridged" output, where neither connects to ground and it will be seriously unhappy if you ground one.
Also, 160 W into 4 ohms is 72 V pk-pk, or 25 V rms. or 36 V pk-pk on each side of a bridged output.  As Benta also stated, a x10 probe will get that down to 7.2 V (0r 3.6 V) pk-pk, which is quite reasonable for an oscilloscope input.

So you guys would recommend that I use the in-line high powered (e.g. >100 watt resistor) to reduce the voltage down to ~25, and using the 10x probe as opposed to higher attenuated probes without inline resistance?

How about using the isolation transformer to plug the amp into as well to be safe?

You're overcomplicating this totally.  I've no idea what you mean by "in-line".

"In-line" as in "in-line resistor."  Not in parallel.  Here is a website that uses the term "inline" to describe a different type of resistor: https://www.jbugs.com/product/9384.html?msclkid=678421e1c7781afdf96a7f0349f65f8d&utm_source=bing&utm_medium=cpc&utm_campaign=**Dynamic%20Search%20Ads&utm_term=jbugs&utm_content=Dynamic%20Website

[bdunham7]

"In-line" as in "in-line resistor."  Not in parallel.  Here is a website that uses the term "inline" to describe a different type of resistor: https://www.jbugs.com/product/9384.html?msclkid=678421e1c7781afdf96a7f0349f65f8d&utm_source=bing&utm_medium=cpc&utm_campaign=**Dynamic%20Search%20Ads&utm_term=jbugs&utm_content=Dynamic%20Website

The terms used in that link only make any sense in the specific case of using a 6 volt original Volkswagen wiper motor on a 12 volt system.  Ironically, that resistor would work as  your speaker load, but will overheat quickly, thus my suggestion use an array of 8R models that look similar.  There's  any number of ways to configure this.

https://www.allelectronics.com/category/530350/resistors/power-resistors-25-100-watts/1.html

However, there's no 'inline' or 'series' or 'parallel' testing an amp--you have a resistor with two terminals and an amp with two terminals.  I suppose you could say connecting the scope across the resistor is 'parallel', but that's not how I would typically explain it.

[gfmucci]

There are some points that aren't clear here.

First, why do you think your amp will put out 126 volts?  That seems unlikely.  Make/model?

Outlaw RR2150.  Because when I put 160 watts and .01 ohms resistance into an ohms/volts/watts calculator I get 126 volts; that's without any "inline resistor" defined as resistor placed in series in the line leading from the amps speaker output to the scope's probe.  Of course, with the 4 ohm load resistor, the volts come down to ~25.

Second, there's absolutely no point in testing an amplifier without a load and I'm not sure what you mean by 'inline resistor'.

I got that piece of advice from Sourav Gupta, Iquesters Solutions LLP, Senior Electronics Design Engineer.  He's the one who suggested no voltage attenuation before the probes, and using a 100x or 1,000x probe.

Typically you use a load resistor similar to the speaker impedance (8 or 4 ohms typically) in place of the speaker.  You could just use a speaker but it would get very loud and their impedance is more complex and variable.  A good first project would be to make a 4/8 ohm load bank.  You need 8 50W 8R power resistors from All Electronics (a couple of bucks each) and  a chunk of aluminum heat sink.

Or these:  https://www.amazon.com/KKmoon-Aluminum-Resistance-Wirewound-Resistor/dp/B082V2N5SR/ref=sr_1_3?dchild=1&keywords=200+watt+8+ohm+resistors&qid=1591907136&sr=8-3

Third, probes are just not an issue.  The standard 1X/10X probes you get with any cheap scope will suffice and in the vast majority of cases, no probes at all would work just fine. The scope has an input impedance of 1M (a million ohms) and many will read +/-40 volts p (80 Vp-p) directly--enough to likely spot your clipping.

  Good to know that.  Thanks. 

As for grounding, don't rely on an isolation transformer unless you clearly know what you are doing.

 
I don't.

An isolation transformer with 3 pronged plugs will still pass thru the ground, which yields you exactly nothing in this case except maybe some blown fuses.

  I got this idea from viewing Dave's "Don't Blow Up Yourself, Your Scope, Your House, and Your Neighborhood" youtube.  That led me to this isolation transformer, with two-prong plug:   https://www.amazon.com/dp/B00B886DGS/?coliid=I2TN4DZI425KYH&colid=3B04RKX85L5W0&psc=1&ref_=lv_ov_lig_dp_it_im

What you need to do is verify that you have a common ground (aka not bridged) amplifier where the negative speaker terminals are tied to ground.  You very likely do have this type of amp, bridged isn't that common. 

  This reminds me of an old Mary Chapin Carpenter song, "I take my chances."  No, no, no.  Not with a $700 amp.  I don't know how to verify that.

 

If not, you have additional complex issues before you scope anything.  In case it hasn't been mentioned yet, PLEASE watch Dave Jones' video "How not to blow up your oscilloscope" until you understand it thoroughly.

I did.  And I'll probably watch it a couple more times before I do anything.

By the way.  The same "blowing up stuff" can occur with power from a USB cable.  Which brings me to this next question:  Will this device work to isolate USB ground, and if not, what will?  https://www.amazon.com/dp/B07GB4CQQ3/?coliid=I3DMV60RNKQJY1&colid=3B04RKX85L5W0&psc=1&ref_=lv_ov_lig_dp_it

[gfmucci]

'Integrated amplifier' sounds consumer. If so, be aware that it probably won't stand running to the clipping point for any significant period of time, especially if you're driving both channels.

I kinda think it will:  https://outlawaudio.com/shop/index.php?id_product=24&rewrite=rr2160-stereo-receiver&controller=product
That's my same amp with same power specs, but with a couple fewer bells and whistles.

[themadhippy]

You very likely do have this type of amp, bridged isn't that common

might not say bridged at the speaker terminals but bridging 2 channel ic amps is fairly  common

[gfmucci]

The terms used in that link only make any sense in the specific case of using a 6 volt original Volkswagen wiper motor on a 12 volt system.  Ironically, that resistor would work as  your speaker load, but will overheat quickly, thus my suggestion use an array of 8R models that look similar.

I used that resistor example noting it was not the same type I would use, but to emphasize that the use of the word "inline" is not uncommon.  It's used to describe resistors, fuses and other components.  Here's one for fuses:  https://www.amazon.com/Inline-Circuit-Breaker-Protection-02-100A/dp/B07628SSWB/ref=sr_1_2_sspa?dchild=1&keywords=inline+fuse+holder&qid=1591908358&sr=8-2-spons&psc=1&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUEySjFHMDMxRFFQRlNHJmVuY3J5cHRlZElkPUEwNzY1ODE4MzQ2OUFRM1FNWVZNTCZlbmNyeXB0ZWRBZElkPUEwMjQ2NDUwMjhDQUZXRkw1VUlCVSZ3aWRnZXROYW1lPXNwX2F0ZiZhY3Rpb249Y2xpY2tSZWRpcmVjdCZkb05vdExvZ0NsaWNrPXRydWU=

This might be a case of using the word "soda" or "pop" in different regions to mean the same thing, although the "soda" guy might not know what "pop" means .

[TimFox]

"Outlaw RR2150.  Because when I put 160 watts and .01 ohms resistance into an ohms/volts/watts calculator I get 126 volts; that's without any "inline resistor" defined as resistor placed in series in the line leading from the amps speaker output to the scope's probe.  Of course, with the 4 ohm load resistor, the volts come down to ~25."

You do not understand the power specification for the amplifier.  If the specification is valid (and there is a long history of manufacturers lying and exaggerating the power), it means that the amplifier will deliver 160 W mean power into a 4 ohm load.  That usually means that it will only deliver 80 W mean power into an 8 ohm load (same voltage, lower current).  These amplifiers are virtually always "constant voltage" sources, so that the output voltage is a weak function of the load resistance, over a reasonable range of voltage (not a dead short.  Therefore, the amplifier will try to put 25 V rms across your dead short, but the current limit (or smoke) will kick in first at the huge current of 25 V into an almost dead short.  Also, if you were to dissipate 160 W in a 0.01 ohm resistor, that would be only 1.265 V rms, not 126.  You should always check the order of magnitude before trusting a detailed computation.

[bdunham7]

Outlaw RR2150.  Because when I put 160 watts and .01 ohms resistance into an ohms/volts/watts calculator I get 126 volts; that's without any "inline resistor" defined as resistor placed in series in the line leading from the amps speaker output to the scope's probe.  Of course, with the 4 ohm load resistor, the volts come down to ~25.

I got that piece of advice from Sourav Gupta, Iquesters Solutions LLP, Senior Electronics Design Engineer.  He's the one who suggested no voltage attenuation before the probes, and using a 100x or 1,000x probe.

I'd suggest you draw a quick schematic of how you intend to connect things and post the picture because there's a lot of room for miscommunication here.  There's no point in a huge semantic debate, but 'inline' will refer to a specific application.  If someone is selling you an 'inline' anything they are selling it for a particular purpose.  In any case, a diagram will make things much clearer.  And I would suggest a 5-amp fast-blow inline fuse on your speaker output(s) (on the red side) while  you are doing this testing.

I got this idea from viewing Dave's "Don't Blow Up Yourself, Your Scope, Your House, and Your Neighborhood" youtube.  That led me to this isolation transformer, with two-prong plug:   https://www.amazon.com/dp/B00B886DGS/?coliid=I2TN4DZI425KYH&colid=3B04RKX85L5W0&psc=1&ref_=lv_ov_lig_dp_it_im

That transformer is not powerful enough to run your amp under full power conditions.  Also, if you have a two-prong cord, the internals are already isolated from the mains plug entirely (but not from the case).

I believe your amp is not bridged.  You will likely have continuity between the two black speaker outputs, but the best way to not blow things up is to start by taking the ground clip off of your scope probe.  Then you can run the amp under power and probe both sides of the outputs, only connecting the ground clip when you are satisfied that there is no signal on the ground side.

[gfmucci]

Here are the amp specs:  110 watts per channel, 20 Hz - 20kHz, 0.05% THD, both channels driven into 8 ohms 165 watts per channel, 20 Hz - 20kHz, 0.05% THD, both channels driven into 4 ohms.

The amp has a 3-prong plug.  Power consumption: 400 watts at full load.
So, something more like this would work?  https://www.amazon.com/Valuepro-GRP-1200-Transformer-Isolation-Benchtop/dp/B00R5CDG8O/ref=pd_aw_sbs_328_2/141-0682910-2872241?_encoding=UTF8&pd_rd_i=B00R5CDG8O&pd_rd_r=8ed81c5d-5c5a-4d7f-87c1-4847de52c860&pd_rd_w=GyOpI&pd_rd_wg=rAg4o&pf_rd_p=c016812c-44e7-4d82-86d0-7b21dc29f2de&pf_rd_r=ZK0PT5ASQXYKW3W4FCXE&psc=1&refRID=ZK0PT5ASQXYKW3W4FCXE

[TimFox]

Those specs make sense, note that the power into 8 ohms is less than 4 ohms.  As written, the spec is not the clipping level, but the power at a reasonable distortion value, and says nothing about shorted load.  The power at 4 ohms is less than twice that into 8 ohms (at stated distortion) due to the higher current required.
Again, I would not assume that the output is non-bridged, since the wrong choice here might destroy the amp.  Bridging is used in solid-state amplifiers to reduce the required DC voltage internally, since it is easier with transistors to run higher current, rather than higher voltage.

[Gyro]

Looking at the manual, it doesn't look like running a full power test would be a good idea - It makes reference to not using two pairs of 6R speakers (ie. 3R load with both switched in) because it will fry the mains transformer (or words to that effect).

If that is an issue that the manufacturer feels worth mentioning in the manual for practical use (ie. listening), then a sustained full power test just doesn't sound worth the risk. If it sounds undistorted at the highest listening level that you are comfortable with, then there's no point is stressing it for the hell of it. Repair might be different.

Industrial and Pro audio amps tend to be designed to take sustained maximum output at minimum load impedance, consumer stuff always has thermal limitations.


P.S. You can forget about the isolation transformer - looking at the rear panel picture, the IEC connector doesn't appear to have a ground pin (unless it's a photoshop glitch).

[gfmucci]

I don't think I want to test that amp any more.  I'll stick with low voltage devices for testing.

[MosherIV]

Hi, if the original point was to test for 'clipping point', you do not actually need a scope.
Your ears will tell  you !
The human ear (brain) is very good at at detecting something wrong with sound.
The sound will start to sound distorted.

However, it is very bad at determining sound level.

[gfmucci]

Hi, if the original point was to test for 'clipping point', you do not actually need a scope.
Your ears will tell  you !
The human ear (brain) is very good at at detecting something wrong with sound.
The sound will start to sound distorted.

However, it is very bad at determining sound level.

I never get the beast to the clipping point.  Too many neighbors would call the cops.  Oh, wait,  the liberal mayors disbanded the PD's.
Seriously, I was wanting to perform this test out of a sense of wanting to learn how to test this sort of equipment properly.  From what folks have been saying, I'd need to spend several hundred dollars on isolation transformers or isolated probes (pricey) and inline resistors (admittedly cheap) so as not to risk blowing up my amp or scope.  So I've abandoned the fling until I learn more.

[Gyro]

I hope we (and I) didn't put you off too much. My comments about consumer don't detract from the quality of the unit for listening pleasure, even at ear-splitting volumes!

I think many of us have done (or know someone who has done) something silly when testing an amp and regretted it! A bit of disappointment now is fairly insignificant compared to the potential for despair and beating yourself up if you kill the thing - especially if the manufacturer refuses to repair it under warranty (fried mains transformer for instance) due to what they deem to be 'abuse'. It's the sort of test that some HiFi mag reviewer might do once on a manufacturer demo unit, but not worth risking your own.