I was looking at my YJ blue board tonight, and noticed this "blemish" on the board. Look at the center of the picture, the area of the board bordered by two inductors and the blue capacitors. Looks like something leaked onto the board. I don't think I was eating while I built this.
Could one of those caps have leaked? Could this be related to the no-input-hiss problem described above?
http://raw-sewage.net/images/diyaudio/tpa3116_blemish_1.jpg
Use a cotton swab to see if it's actually a residue or just a discoloration.
Attachments
so, just received one of the Breeze audio complete 2.1 amps.
Hooked my "sperkers" up to it.
This is being fed by a Nexus 7 tablet running poweramp, all FLAC files. Dayton B-652 speakers with bracing and poly fill, and a crossover designed by Dennis Murphy.
I also have an Optimus 10" passive dual-voice coil subwoofer hooked up in series for 8ohms for the 2.1 set up.
first impressions:
1. it's really small, I cannot get over how compact these amps are.
2. it has a red led indicator light on the front that is useless since it is on any time the power cord is plugged in. so, unless you need to know if it is plugged in, it is kind of a waste since it doesn't let you know if the amp is actually on. I guess this means it is not really wired properly to the switch?
3. speaking of power cords, this one sparks horribly when you connect the power cord. I know sometimes you might get a small arc when you plug in a DC power cable, but this one shoots like a 1 inch yellow spark out of the jack. None of my other class d mini amps exhibit this behaviour. It's very disconcerting and kind of scary.
4. the case is nice, could be a nice piece, but the overall implementation of knobs, the sparky power jack, the way the pots rub against the case, the weird power led, the "sperkers" and cheap silk screened graphics, etc. makes it seem low-end. Hell, it is cheap, but I have $35 amps from Sure Electronics that feature much nicer finish.
5. the board itself looks a little rough. in some places there are tiny smd components jumpered directly to other stuff and connected via micro wires. I thought they were solder blobs, then I realized they were little smd parts. Doesn't look right.
6. the sound ain't bad, but it doesn't seem to play as transparent as the YJ 2.0 black/blue board I have rigged up. With a sub hooked up to the 100w bridged 3116 chip in the Breeze, there is definitely fuller sound, somewhat adjustable, but you have to tweak the satellite volume, the sub volume, and the sub xover freq with the three pots. I guess the middle "FS_Tone" pot adjusts the xover freq for the sub, but it's kind of tough to tell.
7. The Breeze may run warmer than the YJ. I can feel the top of the case getting warm on the Breeze. But, it is sealed up in there.
8. as per other instructions on here, I inspected the board before I used it, I noticed no issue with heat sinks not being attached--but I did notice the aforementioned weird looking jumpered smd parts and such.
So, having spent the morning with the Breeze, I think I prefer the sound of the Dayton b-652s alone with the YJ board as opposed to the Daytons and my Optimus passive 10" sub and the Breeze.
Maybe for near field on the desk the sub is redundant. This would be for mellow stuff, Chris Isaak, Fleetwood Mac, etc. OTOH, if you jam out some house music or techno--like Deep Dish--that sub is pretty nice. But I cannot work at my desk like that.
The Breeze seems okay for $50, but I think there is more potential for the DIY thang with the YJ blue/board. I had a hell of a time just swapping the 1000uF power caps on the YJ for 35v replacements (I run 24v power brick and it was suggested that it's better to provide some more voltage capacity there.)
I don't think there is much I could do to the Breeze given my limited skillsz with the iron.
I am thinking that the way to go would be the YJ board running the satellites and a full on mono bridged 3116 separate for the sub. This little guy definitely works for that, but it's a done deal. I don't think you can do much with it unless you can really do the SMD soldering.
One thing is for sure, I am going broke playing around with all these various 3116 amps. I really do think they are superior to the old smaller tripaths as well as the various TDA749_ series
Hooked my "sperkers" up to it.
This is being fed by a Nexus 7 tablet running poweramp, all FLAC files. Dayton B-652 speakers with bracing and poly fill, and a crossover designed by Dennis Murphy.
I also have an Optimus 10" passive dual-voice coil subwoofer hooked up in series for 8ohms for the 2.1 set up.
first impressions:
1. it's really small, I cannot get over how compact these amps are.
2. it has a red led indicator light on the front that is useless since it is on any time the power cord is plugged in. so, unless you need to know if it is plugged in, it is kind of a waste since it doesn't let you know if the amp is actually on. I guess this means it is not really wired properly to the switch?
3. speaking of power cords, this one sparks horribly when you connect the power cord. I know sometimes you might get a small arc when you plug in a DC power cable, but this one shoots like a 1 inch yellow spark out of the jack. None of my other class d mini amps exhibit this behaviour. It's very disconcerting and kind of scary.
4. the case is nice, could be a nice piece, but the overall implementation of knobs, the sparky power jack, the way the pots rub against the case, the weird power led, the "sperkers" and cheap silk screened graphics, etc. makes it seem low-end. Hell, it is cheap, but I have $35 amps from Sure Electronics that feature much nicer finish.
5. the board itself looks a little rough. in some places there are tiny smd components jumpered directly to other stuff and connected via micro wires. I thought they were solder blobs, then I realized they were little smd parts. Doesn't look right.
6. the sound ain't bad, but it doesn't seem to play as transparent as the YJ 2.0 black/blue board I have rigged up. With a sub hooked up to the 100w bridged 3116 chip in the Breeze, there is definitely fuller sound, somewhat adjustable, but you have to tweak the satellite volume, the sub volume, and the sub xover freq with the three pots. I guess the middle "FS_Tone" pot adjusts the xover freq for the sub, but it's kind of tough to tell.
7. The Breeze may run warmer than the YJ. I can feel the top of the case getting warm on the Breeze. But, it is sealed up in there.
8. as per other instructions on here, I inspected the board before I used it, I noticed no issue with heat sinks not being attached--but I did notice the aforementioned weird looking jumpered smd parts and such.
So, having spent the morning with the Breeze, I think I prefer the sound of the Dayton b-652s alone with the YJ board as opposed to the Daytons and my Optimus passive 10" sub and the Breeze.
Maybe for near field on the desk the sub is redundant. This would be for mellow stuff, Chris Isaak, Fleetwood Mac, etc. OTOH, if you jam out some house music or techno--like Deep Dish--that sub is pretty nice. But I cannot work at my desk like that.
The Breeze seems okay for $50, but I think there is more potential for the DIY thang with the YJ blue/board. I had a hell of a time just swapping the 1000uF power caps on the YJ for 35v replacements (I run 24v power brick and it was suggested that it's better to provide some more voltage capacity there.)
I don't think there is much I could do to the Breeze given my limited skillsz with the iron.
I am thinking that the way to go would be the YJ board running the satellites and a full on mono bridged 3116 separate for the sub. This little guy definitely works for that, but it's a done deal. I don't think you can do much with it unless you can really do the SMD soldering.
One thing is for sure, I am going broke playing around with all these various 3116 amps. I really do think they are superior to the old smaller tripaths as well as the various TDA749_ series
I just switched back to the YJ 2.0 blue board, definitely something different with the gain there. Much louder and definitely more transparent.
I definitely think the hot set up would be YJ board for highs and PBTL mono board for sub with some kind of tweakable crossover freq pot and volume pot.
Maybe one of the heavy hitters on here will do something like that and post the schematic...
I definitely think the hot set up would be YJ board for highs and PBTL mono board for sub with some kind of tweakable crossover freq pot and volume pot.
Maybe one of the heavy hitters on here will do something like that and post the schematic...
I can't design that circuit but know some folks who can. In the meantime, buy a 2x4 miniDSP and you can really have a lot of fun and flexibility. The miniDSP is truly a game changer if you are into 2-way speakers with XO's and don't want to mess with circuits or passive XO's. Coupled with the low cost 3116 amps - it has really changed how I build and listen to speakers.
has anyone checked to see factory PLIMIT setting on any of these amplifiers?
(pin 6 on TPA3116d2, see pp 16-17 of TPA3116 datasheet for examples,
??assume it's disabled if pin 6 open??)
even better, any reports on clipping behavior, whether or not PLIMIT is set?
(by scope, by ear, or by poor little tweeter?,
would be nice to know signal source & gain setting)
(pin 6 on TPA3116d2, see pp 16-17 of TPA3116 datasheet for examples,
??assume it's disabled if pin 6 open??)
even better, any reports on clipping behavior, whether or not PLIMIT is set?
(by scope, by ear, or by poor little tweeter?,
would be nice to know signal source & gain setting)
1.25 meters max speakerwires then? sure3110 disturbed nearby wireless mouses and keyboards without ferrite bead, would the stronger 3116 create similar or bigger problems? maybe further away?? with longer speakerwires I mean, because if you count wires inside speaker cabinet too this would most often leave little to nothing for external speakerwires. Would a second/third set of ferrite beads inside a speaker be an option if wires exceed lenght/problems occur??I also see a ferrite bead output filter
6 PLIMIT I Power limit level adjust. Connect a resistor divider from GVDD to GND to set power limit. Connect directly
to GVDD for no power limit
page 2 datasheet, so not open.
radiosmuck posted nice link in #2424, for me in 3116pbtl I don't think I will reach those levels
sure3110 with volume pot completely open becomes shouty like in screaming, dynamics fade towards flat and it seems louder than it is LOL
to GVDD for no power limit
page 2 datasheet, so not open.
radiosmuck posted nice link in #2424, for me in 3116pbtl I don't think I will reach those levels
sure3110 with volume pot completely open becomes shouty like in screaming, dynamics fade towards flat and it seems louder than it is LOL
Êý×Ö¹¦·Å¿ª¹ØµçÔ´120W 28V/4.2A¹¤Òµ¼¶·ÇÒ»°ãµçÔ´ÊÊÅäÆ÷ÄܱÈ-ÌÔ±¦Íø
The smps from seller small ferrite bead 2.1 amp has 20v-30v range. Any opinions on this?
The smps from seller small ferrite bead 2.1 amp has 20v-30v range. Any opinions on this?
Always plug in the DC plug and after that the mains plug. DC will always spark. DC connectors should be connected in powerless state.
Switches were invented with a reason. Normally a switch is used on the AC side. Problem with switched PSUs for laptops etc is that they don't have AC switches. You could add one for better power-on behavior.
Last edited:
OK. I forgot to add that the contacts of the jack and the plug will develop pits from the arcing and the contact will deteriorate fast. Same counts for relays and switches that switch DC. In short, switching DC is a strain for parts. The quantity of the arcing is dependant on resistance/impedance (capacitance!) of the device that is switched on.
For reliability and safety switching AC is the optimal solution. I hope our friends in China that make these nice electronic devices will understand this one day. Many devices like the Class D amps are without switch so one uses the DC jack as an arcing " switch". If the devices have a switch at all it is often used at the DC side. Sometimes this is done with a small underrated switch soldered on the PCB. It seems like an old almost forgotten bad habit is reviving.
All blah blah. Just remember this. If there is a choice to switch AC or DC: choose AC.
For reliability and safety switching AC is the optimal solution. I hope our friends in China that make these nice electronic devices will understand this one day. Many devices like the Class D amps are without switch so one uses the DC jack as an arcing " switch". If the devices have a switch at all it is often used at the DC side. Sometimes this is done with a small underrated switch soldered on the PCB. It seems like an old almost forgotten bad habit is reviving.
All blah blah. Just remember this. If there is a choice to switch AC or DC: choose AC.
Last edited:
A good non polarized cap across the contacts can often snub a lot of the arcing and was used in automobile engine mechanical contact distributors of yore before we all switched to electronic ignition. Indeed a good "condenser" from an auto supply store for cars is a cheap cap designed just for this purpose. It allowed millions of contact cycles on a DC switch to happen before significant pitting. Give it a try.
Definitely some kind of residue. It didn't come off with just a q-tip, but came completely off with a q-tip and alcohol.
Automotive condensors are big because they handle 1000 v back emf from coils. Try almost anything above 0.47 uF and rated for the vcc with margin.
I think you want non electrolytic caps - polyester film or the like. Something like the little rectangular blue plastic cubes used for the input coupling caps should work. Need to be rated for about 35v though.
Or if you have a radial poly film cap used for a crossover should work too. But I was talking about switch contacts so they go on each side of contact - hard to do if you are plugging in the jack. The jack you will have to deal with by inserting then pluging in the AC wall after. I am talking about to prevent arcing on the actual switch inside the box.
Or if you have a radial poly film cap used for a crossover should work too. But I was talking about switch contacts so they go on each side of contact - hard to do if you are plugging in the jack. The jack you will have to deal with by inserting then pluging in the AC wall after. I am talking about to prevent arcing on the actual switch inside the box.
Last edited:
Sure3110 left out one detail though. The ferrite beads. Although the data sheets says they should be there, they aren't.