Depends how noisy the SMPS is. If your SMPS is a really clean one, then it will make it worse. If your SMPS is horrendous, it will make it better.
These are still switching DC-DC converters, so you get the same type of HF switching noise at the output. The particular one being discussed here has a noise level that I would say is acceptable, very good if you consider the cost of the unit.
These are still switching DC-DC converters, so you get the same type of HF switching noise at the output. The particular one being discussed here has a noise level that I would say is acceptable, very good if you consider the cost of the unit.
If I understand your use case correctly (on the boat, on the beach), you will likely never notice. I completely agree with your sentiments on not overspending for support components. Also, like I said before, if the amp has proper decoupling (and the sanwu 3118 is actually pretty good), even the most noisy supply will be clean at the pins, which for class D is the only place that matters anyway.
I have hooked this DC-DC up to my sanwu 3118 just to see, and my ears do not hear any problems even in a fairly quiet environment. I do fully admit I don't have any sort of special golden ears though, fairly average adult male, maybe even below average.
So in short, I wouldn't worry about it, get all the boards in, hook it up, and only worry about a fix if you notice there is a problem. Maybe I am being overly practical about it, but to be fair, I am technically a mechanical engineer.
I have hooked this DC-DC up to my sanwu 3118 just to see, and my ears do not hear any problems even in a fairly quiet environment. I do fully admit I don't have any sort of special golden ears though, fairly average adult male, maybe even below average.
So in short, I wouldn't worry about it, get all the boards in, hook it up, and only worry about a fix if you notice there is a problem. Maybe I am being overly practical about it, but to be fair, I am technically a mechanical engineer.
Last edited:
300mV - 1.25% at 24V isn't something I would call an issue especially when it's out of audio band.
If there's a problem with this, the usual RC mod at AVCC would help.
On my 3251 boards I have 26mVpp ripple after the 2MHz (42V->15V) buck, which gets down to 15mV after an RC-filter followed by a cheap lm317 to get finally down to 8mVpp. Buck was loaded by 150mA on the other rail (fan). The ripple is further reduced at the pins of the opamp.
Keep in mind that the connection from SMPS to board forms an additional LC-Filter with the help of local bulk decoupling.
If you can't see residual switching noise of the amp on your power rails, all is fine, especially in BTL.
If there's a problem with this, the usual RC mod at AVCC would help.
On my 3251 boards I have 26mVpp ripple after the 2MHz (42V->15V) buck, which gets down to 15mV after an RC-filter followed by a cheap lm317 to get finally down to 8mVpp. Buck was loaded by 150mA on the other rail (fan). The ripple is further reduced at the pins of the opamp.
Keep in mind that the connection from SMPS to board forms an additional LC-Filter with the help of local bulk decoupling.
If you can't see residual switching noise of the amp on your power rails, all is fine, especially in BTL.
Last edited:
Just lost another 3118 mono board!! Pop, pop, pop, pop. I had a pair wired up and leads insulated well so no shorts. What I was doing was attaching and in attaching the positive lead to the battery instead of working a on off switch. If this is the reason for the board looking in a rythmic pattern? Could someone explain why this is any different than flipping an on off switch? Please? I have done this way with countless other amp boards 2497, 3116, etc
Strange. I don't know these boards exactly but maybe they messed up the anti-pop circuit, if any. I had this seen once while testing my own implementation. Magic smoke came out when reconnected in a very short time, like it is when fiddling with power connectors. A switch does not bounce that much when doing the connection.
This is only valid if an external antipop is on the board, can't tell.
This is only valid if an external antipop is on the board, can't tell.
hmm, what power supply are you using right now, and how is it wired up?
How about speakers, what impedance are they, how many, and how are they wired?
Sorry if this has been covered before, this thread has been busy lately.
Mord, don't know if this will help you determine anti pop but here is the best photo I can get for right now. I will look at the datasheet again later.
How about speakers, what impedance are they, how many, and how are they wired?
Sorry if this has been covered before, this thread has been busy lately.
Mord, don't know if this will help you determine anti pop but here is the best photo I can get for right now. I will look at the datasheet again later.
Attachments
Last edited:
Problem fixed!!!!after pouting, getting all moody, I decided to wire up the mute function on the last working board to see if I could get it to last longer? While I was at it I hooked up a switch to the piping amp too, why not strick while the iron is hot! And ,Shazam thank you Mam! The poping has stopped!!!! I also feel it's a little louder, the red LED is brighter- or just maybe a result of my jubilation to continue tomorrow with the build and not have to wait on a slow boat from China?
So, apparently my Sanwu 3118 boards need the mute wired up! Maybe Doc or Jensen could look at the way the mute function is routed on these boards? I flipped the switch off ànd on several times to be sure my positive results were repeatable, so yeah!
So, apparently my Sanwu 3118 boards need the mute wired up! Maybe Doc or Jensen could look at the way the mute function is routed on these boards? I flipped the switch off ànd on several times to be sure my positive results were repeatable, so yeah!
When you say the mute is wired up, what do you mean exactly? It looks like they have a 100k pull down on the mute pin, and the adjacent pin on the jumper is connected to GVDD with a 10kR in series. Should NOT be muted by default, and I wouldn't advise using that jumper to mute it!
That said, whatever it is you did, may as well try it on the others to see if they can be revived. If it's already broken, what is there to lose?
Here is a photo with much better traces shown. Using my signature method for reverse engineering PCBs, a really bright flashlight, enjoy
EDIT: For the picture worth noting, the only thing on the back side is ground plane, so all vias are to ground.
That said, whatever it is you did, may as well try it on the others to see if they can be revived. If it's already broken, what is there to lose?
Here is a photo with much better traces shown. Using my signature method for reverse engineering PCBs, a really bright flashlight, enjoy
EDIT: For the picture worth noting, the only thing on the back side is ground plane, so all vias are to ground.
Attachments
Last edited:
