Power Supply Whine

[wwenze]

Hi all, mechanical engineer here building a speaker and I am hoping to get some clarification on my design. Currently, my speaker plays music, but there are a few issues...

Overview of my active, powered speaker build components:

1. Peerless by Tymphany SDS-P830657
2. Dayton Audio ND25FW-4 1" Tweeter
3. LRS 150 24 PSU
4. XKitz LR 2 Way Active Crossover
5. Amp, Wondom AA-AB32174

My questions:
1. You can see this in the attached video file, but there is a horrible whine that plays when the speaker turns on. I powered the system using a desktop power supply at work, and there was no whine at all, so I am sure it's coming from the power supply. Im hoping to get some EE explanation on where this comes from and what I can look for in other power supplies to fix this problem.

2. Currently, I am just feeding in a RCA audio source from my record player. There is no volume control on the speaker itself, but I would like to add that. Can I just take a Potentiometer and smack it in-between the RCA input and the RCA input to the crossover? I have the hardware to do that, I just wasn't sure if that's valid or if it will introduce some similar noise issues.

Video Link

Looking forward to chatting about this!

Edits:
Added Amplifier to component list

Is the XKitz LR 2 Way Active Crossover also being powered by the Meanwell? I see two sets of cables coming out from the PSU.

Care must be taken when powering multiple analogue equipment from the same SMPS, especially when they have a different "mid point" voltage.

Without going into details and without spending time troubleshooting, the solution is to get a power supply that works with this application.

Alternatively you can just power the XKitz with its own power supply just like what happens when you buy separate products, and see if the problem goes away.

Add: Instead of connecting the XKitz directly to the Meanwell, you can also try to tap the voltage from the PCB of the amp, often such amp modules will have locations for you to daisy chain the power cabling which is useful for situations like this. It won't solve the fundamental problem behind, but if you're lucky you may reduce the noise to the same level as a soundcard in a computer.

This module reminds me of Sure boards for some reason. I used the included barrel cable to connect to the PSU, which leaves the screw terminals available for other stuff.

 

[tanno1]

Is the XKitz LR 2 Way Active Crossover also being powered by the Meanwell? I see two sets of cables coming out from the PSU.

Care must be taken when powering multiple analogue equipment from the same SMPS, especially when they have a different "mid point" voltage.

Without going into details and without spending time troubleshooting, the solution is to get a power supply that works with this application.

Alternatively you can just power the XKitz with its own power supply just like what happens when you buy separate products, and see if the problem goes away.

Add: Instead of connecting the XKitz directly to the Meanwell, you can also try to tap the voltage from the PCB of the amp, often such amp modules will have locations for you to daisy chain the power cabling which is useful for situations like this. It won't solve the fundamental problem behind, but if you're lucky you may reduce the noise to the same level as a soundcard in a computer.

View attachment 507764

This module reminds me of Sure boards for some reason. I used the included barrel cable to connect to the PSU, which leaves the screw terminals available for other stuff.

Hi wwenze, yes the PSU is powering both the amp and the crossover. I will try the suggestions you mentioned.

 

[john61ct]

Meanwell are supposed to be on the quality side

Famous enough that fakes are sold.

 

[john61ct]

Please post your purchase link for the PSU.

Have you adjusted the output V/amps at all?

4. XKitz LR 2 Way Active Crossover

Very interested in this xover, anyone recommend a suitable enclosure, or cheap custom maker?

 

[tanno1]

@john61ct Link to PSU

As for the xo enclosure, I have made a CAD enclosure. I can send you a STEP file to print or print you a copy when I print them.

 

[john61ct]

SUPER!

ofc posting it somewhere for posterity would be Even Better

 

[tanno1]

Awesome. Ill print, test, and possible need to refine it but then I will post it in this thread. @john61ct

 

[john61ct]

So that will be genuine Mean Well from Digi.

If your other troubleshooting doesn't help, try their HRP series, ideally fanless

 

[Sokel]

Please post your purchase link for the PSU.

Have you adjusted the output V/amps at all?

Very interested in this xover, anyone recommend a suitable enclosure, or cheap custom maker?

Its specs don't make sense to me, they state an about 80dB THD+N.
Unless it's about the whole 20Hz-20KHz range, it's extremely poor, cheaper ones do better.

 

[tanno1]

@wwenze This Video shows me powering the XO from a Sigilent SPD3303X-E, the Amp from the LRS-150-24, and playing sound well.

When I power the XO and the amp from the LRS-150-24, I get the same hideous whine seen here

If that is the case, I believe I need to get a second power supply. Maybe I can downsize the LRS-150 and then I would bridge the AC input from one to the other.

I don't see any suitable places to ground a wire between the amp and the XO as someone else previously mentioned.

 

[Wolf11Man]

Try this: Remove the ground wire to the x-over. Now it is grounded by the RCA connection. Does the whine change or disapear?
This would point to a ground problem.
Power the amp and the x-over from your Siggelent PSU. If sound disapears, this points to the SMPS you use.
You got to isolate such a problem, this is part of any amplifier including electronic build.

 

[john61ct]

Chassis to chassis grounding.

Ideally only one device in the chain actually Earthed, hopefully the only one with three prongs, and the outlet installed properly, central circuit box to the copper rod pounded down below water-saturation (even in summer) point.

While researching going balanced inputs to GFA-555, came across "Hoppe's Brain Earth-Loop Breaker board"

Neutrik's "combo" receptacles -- combined female XLR & female 1/4" connectors -- piercing tab feature tie to chassis ground

and also the "Muncy Solution"

Bill Whitlock paper, dedicated to Neil Muncy

https://indianaaes.org/wp-content/uploads/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf

"The Absolutely Best Right Way to Do It" from Rane note 151

Grounding and Shielding Audio Devices

Since the Audio Engineering Society has adopted the "pin 2 is hot" standard, what is the correct use for pin 1?

www.ranecommercial.com

All this is WAY above my pay grade, so if there any conflicts between any of the above, please point them out.

 

[tanno1]

Try this: Remove the ground wire to the x-over. Now it is grounded by the RCA connection. Does the whine change or disapear?
This would point to a ground problem.
Power the amp and the x-over from your Siggelent PSU. If sound disapears, this points to the SMPS you use.
You got to isolate such a problem, this is part of any amplifier including electronic build.

Video showing me removing the grounding wire from the XO. That did not fix the problem, it maybe made it a bit quieter.

I have powered the XO and the amp from the Sigilent and the problem was fixed. I believe this may be an issue with my power supply then, maybe I need to replace it with the same one or a smaller one.

 

[tanno1]

Hi all,

Attached is a new schematic sketch I have came up with to power the XO independently of the LRS based on feedback from this post. I found the12V 2.5A IRM-30-12ST, an AC-DC power supply. The output is 12V, 2.5A with 150mVp-p, within the range of powering the XO. From the XO data sheet, it says that it has an "Filtered LDO regulator for high power supply noise rejection", so I am assuming this would take care of the 150mVp-p noise.

If anyone has any feedback, please let me know what you think. Once I get this sorted I will implement the volume control. I have some photos of the speaker to share but I think I will wait till I finish up this power supply issue.

 

[solderdude]

That would be the way to go most likely.

 

[john61ct]

Potted does not necessarily mean quiet?

But let us know how it goes.

EPS-45-12, RS-35-12 are alternatives

 

[solderdude]

An alternative could even be a 24V//24V DC-DC converter as long as it is isolated.
Let's hope that the powersupply used has very little leakage current (which isn't always the case with SMPS).
the ones intended for medical usage usually are very low in leakage.
Like this GSM18 Meanwell or XP Power ACM06 for instance

 

[tanno1]

Update video

Whine has reduced a ton, but the overall sound isn't great currently. On the XO, there is a Baffle Step Compensation, Treble, and Bass pot that can all be adjusted.

Before I go digging into audio testing, does anything have any experience or know exactly where I can go to find a good guide on basics of testing my speaker as I incrementally tweak it?

I have an audio technical AT-2035 Mic that I figured I might be able to use, but not sure how that exactly works in tandem with software to test and characterize a speaker.

 

[Sokel]

Update video

Whine has reduced a ton, but the overall sound isn't great currently. On the XO, there is a Baffle Step Compensation, Treble, and Bass pot that can all be adjusted.

Before I go digging into audio testing, does anything have any experience or know exactly where I can go to find a good guide on basics of testing my speaker as I incrementally tweak it?

I have an audio technical AT-2035 Mic that I figured I might be able to use, but not sure how that exactly works in tandem with software to test and characterize a speaker.

Instead of starting measuring speaker response right from the start, I would start with some electrical sanity checks at the outputs of the x-over.
Like separate output voltage for different ways, slopes, FR, noise, etc.

If not with an ADC (audio interface, etc) and REW or Multitone Analyzer, at least with a DMM.