EEVBlog Rebranded Brymen BM235 & Backlight noise research

[Scottjd]

So I've had a couple of nights without sleep, or minimal sleep at most.
But every time I woke up one question was on my mind "What is that noise with the BM235 back light?"
Oh, and the fact that Dave's Manual said it would NOT void the warranty. That alone was a big tease and I just had to take it apart.

So I know this video is not the best quality. And heck I might even be a little goofy  from lack of sleep, I just needed to open that meter  and look around to see what made that high  pitch noise.

So I did not originally intend to do this as a video, it was more of a after thought and this is why the video starts with the meter back already pulled off. But I explain the things you need to watch out for and any surprises you may run into.
But I did record putting it back together and the follow up testing to make sure it was still as accurate just as it was when I did the review. Since I have the numbers from my the review I can say the tear down did not affect the meter at all.
I also ruled out some possibilities for the noise, and think I may have narrowed down whats causing it.
Oh, and I know the lighting is bad. I tend to forget things without sleep like turning on the lights.... 

Its time for me to get some more  , I hope...... Enjoy the video.......

If your interested in the review this is the thread:
https://www.eevblog.com/forum/testgear/eevblog-branded-brymen-bm235-in-depth-feature-review-and-thoughts/

https://youtu.be/mcz8Ovm5NmQ

[idpromnut]

I watched your video, nice work   Although I was yelling at the screen a bit when you were saying that the whining was coming from the LEDs

After a bit of hunting, it looks like if you remove C44 (smoothing cap on the backlight charge-pump output) the whine will be completely gone (except if you can hear into the 32kHz range, then you're screwed as that is the base frequency that the charge pump is running at).  I don't see any long term issues with it running like that, and there are no parts to add.

C44 is a ~ 10uF ceramic.

A few current measurements (using a uCurrent + Fluke 27):

With C44 in place:
- No backlight on:  2.2mA
- With backlight on: 26.2mA

With C44 removed:
- No backlight on:  2.2mA
- With backlight on: 27.3mA

[idpromnut]

I also shot a couple of videos of me bumbling around trying to find the cause:



And finally the fix:

[Brumby]

I wonder if we can get an 'official' thumbs up or thumbs down on this - or an alternative?

[MarvinTheMartian]

Probably a very, very stupid question (but hey, I'm new)  ...

Would some snot glue on the cap dampen down the vibration at all? 

(I'm assuming it is a physical vibration of the cap.)

I probably don't have a clue what I'm talking about - right?  Doh!

[EEVblog]

I wonder if we can get an 'official' thumbs up or thumbs down on this - or an alternative?

Brymen are looking into it.
Changing the cap to another brand, model or size etc will likely make a difference.
MLCC capacitor microphonics are very manufacturing process specific.

[wraper]

If the original cap is Z5U or similar, changing it to X5R, X7R type most likely will kill the noise.

[ebastler]

Changing the cap to another brand, model or size etc will likely make a difference.
MLCC capacitor microphonics are very manufacturing process specific.

Is there an accepted term for the opposite of "microphonic" behavior, as present in this case?  "Speakerphonic"? 

Anyway, it's not an uncommon effect. Murata are even promoting "low noise capacitors" to fix it:
http://www.murata.com/products/capacitor/mlcc/solution/Naki

[EEVblog]

Is there an accepted term for the opposite of "microphonic" behavior, as present in this case?  "Speakerphonic"? 

Reverse piezoelectric effect among others.
I mentioned it back in episode 33

[Salas]

Changing the cap to another brand, model or size etc will likely make a difference.
MLCC capacitor microphonics are very manufacturing process specific.

Is there an accepted term for the opposite of "microphonic" behavior, as present in this case?  "Speakerphonic"? 

Anyway, it's not an uncommon effect. Murata are even promoting "low noise capacitors" to fix it:
http://www.murata.com/products/capacitor/mlcc/solution/Naki

The "Microphonic" old term comes from the reasoning that a component picks up a frequency mechanically. More accurate would be "resonant" in the audio band.

[wraper]

More accurate would be "resonant" in the audio band.

Completely inaccurate.
You can call it "sound emission", "resonant in audio band" means what it says (resonant, by the nature of the part, electrically resonant), nothing about vibration/emitting the sound.
http://www.keysight.com/upload/cmc_upload/All/exp79.pdf

[ebastler]

Is there an accepted term for the opposite of "microphonic" behavior, as present in this case?  "Speakerphonic"? 

Reverse piezoelectric effect among others.
I mentioned it back in episode 33

Yep, I realize it's mostly a piezoelectric effect in capacitors. Other effects might inadvertently translate electrical signals into sound as well (in other components): Electromagnetic force of course is a common source of transformer hum. Electrostatic forces, maybe, in some types of capacitors? Acoustical noise from modulated light pressure optical data transmission, anyone? (Very high ultrasonic pitch, hence people don't often complain about it )

Just meant to say that "microphonics" did not seem the right term here, and that I was surprised to realize that the inverse effect does not seem to have a general name -- at least not one that is commonly used.

[Salas]

More accurate would be "resonant" in the audio band.

Completely inaccurate.
You can call it "sound emission", "resonant in audio band" means what it says (resonant, by the nature of the part, electrically resonant), nothing about vibration/emitting the sound.
http://www.keysight.com/upload/cmc_upload/All/exp79.pdf

Why only electrically resonant, when it can also mean mechanically resonant or piezoelectrically resonant just as well?

[iamdarkyoshi]

I wonder if we can get an 'official' thumbs up or thumbs down on this - or an alternative?

Brymen are looking into it.
Changing the cap to another brand, model or size etc will likely make a difference.
MLCC capacitor microphonics are very manufacturing process specific.

Is it physically the cap making the noise? Because I remember that Bigclive said that ceramic caps tend to have piezoelectric qualities that cause them to make slight noises when they get charged/discharged

[Kleinstein]

The light pressure is quite low, so you usually will not hear this. Another source of sound is due to thermal effects - AC current causes modulation in temperature and thus through thermal expansion a mechanical vibration. This works with both thermal expansion in the solids and air. It's usually not very loud unless amplifies by a resonance, but it is measurable .At university I have measured this effect on solar cells when passing AC current through them - this even allows to distinguish where exactly the diode gets hot. It funny where exactly a diode / solar cell gets hot and cold.

The piezo effect with capacitors can vary a lot, as this is a second order effect and needs some asymmetry in polarity to have a piezo effect. There is a small quadratic electrostriction effect that is present in essentially all materials, not just the high capacitance MLCC.

[marshallh]

I will swap out the cap and see what happens.
This is also happened to me on a switching reg with MLCCs.

[wraper]

Why only electrically resonant, when it can also mean mechanically resonant or piezoelectrically resonant just as well?

But it is not mechanically resonant at the frequency you are hearing the whine.

[markone]

Why only electrically resonant, when it can also mean mechanically resonant or piezoelectrically resonant just as well?

But it is not mechanically resonant at the frequency you are hearing the whine.

How can you exclude that ?

It could be electrical, mechanical or a combination of both.

I guess that capacitor is there to low pass and level the LEDv PWM driver's output, in par with the inductance of course, so it would be interesting to know L & C values a put a scope's probe to see if the resonace is actually nearby the square wave's fundamental or is from edges overshoots.   

[MarvinTheMartian]

Wow!

This sounds a whole lot more complicated than I had assumed.

I thought it would have to be a purely physical vibration to produce sound.

Must say I didn't notice the whine in my EEVblog meter until read this topic.

I have to hold the meter about 12cm from my ear before I hear the whine (and I have almost perfect hearing according to my hearing test yesterday).

Maybe I was lucky with the meter Dave sent to me - whoo-hoo! 

Edit: Just found out I have a slight dip in my hearing around 4Khz (still in normal range though) so that might explain my not easily hearing the the whine at 5KHz.

[wraper]

Why only electrically resonant, when it can also mean mechanically resonant or piezoelectrically resonant just as well?

But it is not mechanically resonant at the frequency you are hearing the whine.

How can you exclude that ?

It's whining at the electric ripple frequency. Good luck to get it to match exactly with a capacitor mechanical resonance frequency. So no, you cannot call the capacitor whine being resonance. Think about speaker or piezo buzzer, yes they have resonance frequency(ies) but they do emit the sound in much wider frequency range.

[Salas]

It can be excited and sing in whatever mechanical frequency it has. When you hit a bell even once you do not hit with its resonant frequency, you just excite it.

[wraper]

It can be excited and sing in whatever mechanical frequency it has. When you hit a bell even once you do not hit with its resonant frequency, you just excite it.

Then please demonstrate how to excite the piezoelectric transducer by applying some different frequency signal to it in a way that it will emit the sound with it's own resonant frequency.

[markone]

It's whining at the electric ripple frequency. Good luck to get it to match exactly with a capacitor mechanical resonance frequency. So no, you cannot call the capacitor whine being resonance. Think about speaker or piezo buzzer, yes they have resonance frequency(ies) but they do emit the sound in much wider frequency range.

With XTALs what you call ripple frequency is in perfect phase with the physical device's mechanical resonance tone, also capacitors with piezo effects have mechanical resonant frequencies.

Let's put things in this way : the whining tone could be the result of step response of a resonant system with a mechanical component, with have a square wave so lot of edges 

I'm not saying is like that for sure, i'm saying it could be, without any data we can argue for days without converging.

So it's time to put a scope's probe on the darn thing and post the waveform 

[markone]

It can be excited and sing in whatever mechanical frequency it has. When you hit a bell even once you do not hit with its resonant frequency, you just excite it.

Then please demonstrate how to excite the piezoelectric transducer by applying some different frequency signal to it in a way that it will emit the sound with it's own resonant frequency.

It's called resonant system step responce, if the piezo transducer  has a one or more resonant points (and everything has at least one), applying a low freq square wave, maybe under its freq responce,  you could obtain high freq tones on par with resonant points.

[markone]

Just one question to the OP: the whining noise's level is so high to be a deal breaker ?