There's a simpler mod to get a power-on mute circuit. I had it shown on the 3116 thread.
Would be interesting where your "fatter" bass comes from, either from additional capacitance or from your imagination.
If first, I'd check the power line with a scope to see, if if the voltage is sagging the amount that results in "weak" bass as bandwidth isn't changed by bulk capacitance.
Lower valued pot should help with hiss. Is there also hiss/noise with no source connected?
Would be interesting where your "fatter" bass comes from, either from additional capacitance or from your imagination.
If first, I'd check the power line with a scope to see, if if the voltage is sagging the amount that results in "weak" bass as bandwidth isn't changed by bulk capacitance.Lower valued pot should help with hiss. Is there also hiss/noise with no source connected?
Yes, I know - I've implemented it on my Breeze amps.
WIll try to find out the proper soldering points on the Sanwu. But, man, those SMD components are tiny
I'm not sure my soldering skills (and iron) are up to the task...
doctormord said:Would be interesting where your "fatter" bass comes from, either from additional capacitance or from your imagination.
Could be just my imagination/expectation, I admit
However, there is no bass lacking now... it definitely has enough bass.
For testing bass response I usually use "American Beauty" motion picture score by Thomas Newman - there's a lot of "synthetic" deep bass in places, and I can easily hear how an amp behaves in those passages...
doctormord said:
Yes, there is hiss with no source connected....
I've measured my Raspberry Pi source signal voltage, and it definitely goes too high - above 1V sometimes, so I guess lowering the pot value should help - I'll try both 10K and 20K pots, and see which behaves better....
Need to get some appropriate pots first...
My speculation as to what is happening is you're getting high frequency capacitive pickup on your wires to the pot. When the pot is in the middle that's when its output impedance is highest, meaning most susceptibility to noise.
What I would do is put in an RC filter right at the input to your TPA3118. The value of the R would need to be somewhere in the kohm range (3k3 to 10k) and the C about 100pF. The R is in series with the signal, the C goes to 0V from the input pin of the TPA3118.
possible anti-pop (?)
Here's what I was able to deduce - please, correct me if I'm wrong (and sorry for the poor drawing, not my forte...).
In short, a "Giancarlo/Drmord" anti-pop solution, from what I was able to deduce from the board and traces, with a little help of a multimeter...
But, the soldering (if there is to be any at all, still not sure whether I'll go on with it) will be tricky: even if I forego the Schottky barrier diode on R30, I still have to solder one end of the 150K resistor on R23... very tight
Anyway, please let me know if I got this right... It's late and I'm not thinking straight...
Thx!
PS: I got hold of a 10K log stereo pot, and it helped - the "hiss area" on the pot moved to high volume side, it's a bit narrower, and in general leaves me a lot more "usable" volume area (if we take 6 o'clock to be max. volume, the hiss now appears between 3:30 and 5:00, approximately).
But, if I'm going to use these as pure monoblocks (separate enclosures, without potentiometer), I guess it will be a non-issue....
Here's what I was able to deduce - please, correct me if I'm wrong (and sorry for the poor drawing, not my forte...).
In short, a "Giancarlo/Drmord" anti-pop solution, from what I was able to deduce from the board and traces, with a little help of a multimeter...
But, the soldering (if there is to be any at all, still not sure whether I'll go on with it) will be tricky: even if I forego the Schottky barrier diode on R30, I still have to solder one end of the 150K resistor on R23... very tight
Anyway, please let me know if I got this right... It's late and I'm not thinking straight...
Thx!
PS: I got hold of a 10K log stereo pot, and it helped - the "hiss area" on the pot moved to high volume side, it's a bit narrower, and in general leaves me a lot more "usable" volume area (if we take 6 o'clock to be max. volume, the hiss now appears between 3:30 and 5:00, approximately).
But, if I'm going to use these as pure monoblocks (separate enclosures, without potentiometer), I guess it will be a non-issue....
Attachments
I wouldn't go for this "big" 10uF, better is higher resistance and lower value capacitor - in case of component size. (It's also better to have less capacitance when you deside to left the diode out.) I don't know if the "Giancarlo" mod is needing 10uF.
You may go this way:
Diode is not "a must", but nice to have.
But all in all, doable.
You may go this way:
Diode is not "a must", but nice to have.
But all in all, doable.
Last edited:
The diode protects the MUTE-input from reverse voltage/current which will be there if you power-off the device. (As long as the capacitor is discharging) You can rely on the internal protection diode, 1uF isn't that much energy but to be really safe, you'll put this diode in parallel. Saying so, this is only tested with 1uF - if you go for higher capacitance you might damage the internal protection diode.
Heh...
I have my Sanwu 8932 boards plugged directly to speaker terminals with 1in long transformer wire. Theyre powered by linear unregulated 29vdc supply. Sounds very good with impressive vocal dynamics.
Am itching to try the 3118, but maybe I'll wait for you to report how your 8932 compares.
Heh...
I have my Sanwu 8932 boards plugged directly to speaker terminals with 1in long transformer wire. Theyre powered by linear unregulated 29vdc supply. Sounds very good with impressive vocal dynamics.
..and don't forget the orgasmic bass that the 8932 produces!!!
Man, I hate soldering such tiny things
My eyesight isn't what it used to be (to say the least), and I look positively hilarious with 2 pairs of reading glasses stacked on top of my nose
Seriously, this Sanwu TPA3118 is tiny... Besides, looks like the components aren't really "top notch"... Looks like they used substandard smd resistors - I've had two die on me during soldering... The contacts/pads on the ends of resistors were rather deteriorated... Had to resort to some creative microsurgery - the SMD resistors I have at hand are two sizes too big, so I had to solder them upright - and I managed to also remove one of the pads on the board
So, the other end (where I lost the pad) was then soldered directly to pin 2 (actually, between legs 2 and 3 - SDZ and FAULTZ)...
Talk about microsurgery.... phew!
One board/channel now behaves as expected, with antipop implemented properly, but the other one is quite strange - it starts playing only when I turn it OFF!
Strange - I used the same components, same soldering points (apart from that hack with SMD resistor mounted upright).... The left channel behaves as expected - starts playing after about 1 sec. delay. The rigt channel/board remains silent. But, when I cut off the power, I can hear about 1/2 second of sound from it - as the capacitor is draining
In short, these boards are turning out to be a DIY nightmare (for an amateur with a soldering iron like me)....
More investigation to follow tomorrow - I've had enough for today...
My eyesight isn't what it used to be (to say the least), and I look positively hilarious with 2 pairs of reading glasses stacked on top of my nose
Seriously, this Sanwu TPA3118 is tiny... Besides, looks like the components aren't really "top notch"... Looks like they used substandard smd resistors - I've had two die on me during soldering... The contacts/pads on the ends of resistors were rather deteriorated... Had to resort to some creative microsurgery - the SMD resistors I have at hand are two sizes too big, so I had to solder them upright - and I managed to also remove one of the pads on the board
So, the other end (where I lost the pad) was then soldered directly to pin 2 (actually, between legs 2 and 3 - SDZ and FAULTZ)...
Talk about microsurgery.... phew!
One board/channel now behaves as expected, with antipop implemented properly, but the other one is quite strange - it starts playing only when I turn it OFF!
Strange - I used the same components, same soldering points (apart from that hack with SMD resistor mounted upright).... The left channel behaves as expected - starts playing after about 1 sec. delay. The rigt channel/board remains silent. But, when I cut off the power, I can hear about 1/2 second of sound from it - as the capacitor is draining
In short, these boards are turning out to be a DIY nightmare (for an amateur with a soldering iron like me)....
More investigation to follow tomorrow - I've had enough for today...
You need lighted binocular magnifying goggles, good Swiss made tweezers, a Panavise, good 0.015in soldering iron tip, steady hands, and a bit of practice. Find an old dead PC MB of peripheral card and try desoldering and resoldering.
Turns off to turn on? Strange.
But my Sanwu's don't have pop problem. A very soft almost inaudible "click" sound. Sharp not a thump.
Turns off to turn on? Strange.
But my Sanwu's don't have pop problem. A very soft almost inaudible "click" sound. Sharp not a thump.
xrk,
I am thinking of getting this board soon (may be those 8932 too
). A quick question - the inductors on this board is 10 uH, so looks like the filters are configured for 4 ohm speakers, are you using 4 ohm speakers? Also, the 60 W power that Sanwu claimed is for 8 ohm speakers? I recalled the data sheet said 3118 does 30 W for 8 ohms stereo (correct me if I am wrong) in BTL. Since this one is in PBTL therefore 1 x 60 W. Thanks!Regards,
xrk,
I am thinking of getting this board soon (may be those 8932 too
Regards,
I tested with a 5ohm dummy load - so that could be called an 8ohm speaker as some get down to 5.7ohms Re. I got a solid 60w and 75w if momentarily pushed. Don't know what the THD would be but thermally it is able to drive enough current to produce the wattage with a 24v supply.
Generally a 30w at 8ohms BTL can only achieve 60w PBTL if impedance goes to 4ohms.
Thanks for the suggestions - but I had to make do with what I have...
A better set of eyes would come handy, too
Anyway, I managed to get this in order - both boards now have properly implemented "anti-pop" with about 1 second delay after turn-on.
The problem with the other board (which played only for 1/2 sec. when turned OFF) was that the 100K resistor leading to the mute pin on the chip had a bad contact. A quick touch with the soldering iron on both ends got it functioning properly again
Am now enjoing the sound of "dual mono" system. The boards sound nice. I also added 1500 uF capacitor to each board, just in case...
In short, I'm happy with these, so I ordered some more...
BTW, can anyone recommend a solid (and small) tube preamp to be used with these (from ebay, as usual
)?Ah, forgot to reply to this, although you probably found out for yourself:
= yes, just short those two contacts for mute.
Actually, you can wire a simple on/off switch if you want to get fancy and avoid handling paperclips
Ah, forgot to reply to this, although you probably found out for yourself:
= yes, just short those two contacts for mute.
Actually, you can wire a simple on/off switch if you want to get fancy and avoid handling paperclips
I think that is how the push button "on/off" switch on most amps for the 311x work. Amp is basically either on or in standby.
I received a pair of Sanwu TPA 3118's from China today. Guess I beat the Chinese New Year shut down. They came pretty quick. Ordered Jan. 19th.
Wow! these are really small. They are smaller than the small sized Post-it-Note. 45X35MM board size. It's difficult to see the components without magnification. I checked every solder connection using a jewelers eye loop. Everything looks good, and the boards are really clean.
Having never worked with mono amps, I have a few questions: Should I treat these as 2 separate amps, one powering each channel with 2 power supplies, or can I use one PS. I have two 24V SMPS but they are not exactly alike. One is outputting 24V, the other 24.5V When using 2 separate amps (mono blocks) should there be any connection between the 2 amps (grounds?).
Any tips on implementing these tiny amps...greatly appreciated!
Wow! these are really small. They are smaller than the small sized Post-it-Note. 45X35MM board size. It's difficult to see the components without magnification. I checked every solder connection using a jewelers eye loop. Everything looks good, and the boards are really clean.
Having never worked with mono amps, I have a few questions: Should I treat these as 2 separate amps, one powering each channel with 2 power supplies, or can I use one PS. I have two 24V SMPS but they are not exactly alike. One is outputting 24V, the other 24.5V When using 2 separate amps (mono blocks) should there be any connection between the 2 amps (grounds?).
Any tips on implementing these tiny amps...greatly appreciated!