hi
but if you have this issue with all your TPA3116 boards its unlikely that all are "defect".
so maybe your psu still need a closer look- weak psu, cabeling?
i do not think that your sub gets lower than 2ohms? TPA3116 can handle 2R in mono configuration (pbtl) if i am right. but maybe the 4R sub draw too much current and therefore the ocp-over current protection shut down
datasheet chapter 6.3 + 6.6 (4.5Amps, 7.5amps)
speakers
i guess your sub is not in a closed housing because if you "overdamped" your inserts in the sub in a closed hosuing.... you might get too less possible movement at your sub and therefore the sub can´t move and the coil is "like" a DC resistor and draw high current
chris
chris
I tested with a lot of different speakers.
8ohm, 4ohm and 3ohm.
Also when only pluged the 8ohm satellite speaker i get same problem.
In my last project subwoofer is bassreflex and 3ohm. Medium and tweeters are for each channel only 2ohm :-/
But when I change to 8ohm satellite same problem.
How can I mesure output current?
My multimeter is to slow I think. I have an oscilloscope. Shoul I put a shunt and mesure the voltage drop?
Thanks for your help
I tested with a lot of different speakers.
8ohm, 4ohm and 3ohm.
Also when only pluged the 8ohm satellite speaker i get same problem.
In my last project subwoofer is bassreflex and 3ohm. Medium and tweeters are for each channel only 2ohm :-/
But when I change to 8ohm satellite same problem.
How can I mesure output current?
My multimeter is to slow I think. I have an oscilloscope. Shoul I put a shunt and mesure the voltage drop?
Thanks for your help
hi
this 8R satelite are really 8ohms? this reminds me at my Kef Q100 its written 8R but i measure new 4R DC and in the net i found some graphs with 4R impedance...
...so really check your speakers impedance. with DMM you can measure the DC of you speakers ( no cabling, amp etc) and give you a rough overview what you have.
current
a shunt sould be as low as possible to 0,1R so if you have 4,5 amp and more you get U=i*R= 0,45 volts for more than some seconds. practically it could be that between you speakers output terminal -cable and your speakers plug you still have 0,1R "loss"....
better:
you should measure at the pin of the TPA chip where you should get the ocp or mute command for the chip.
chris
this 8R satelite are really 8ohms? this reminds me at my Kef Q100 its written 8R but i measure new 4R DC and in the net i found some graphs with 4R impedance...
...so really check your speakers impedance. with DMM you can measure the DC of you speakers ( no cabling, amp etc) and give you a rough overview what you have.current
a shunt sould be as low as possible to 0,1R so if you have 4,5 amp and more you get U=i*R= 0,45 volts for more than some seconds. practically it could be that between you speakers output terminal -cable and your speakers plug you still have 0,1R "loss"....
better:
you should measure at the pin of the TPA chip where you should get the ocp or mute command for the chip.
chris
So I change to tas5611 and no more problems.
Just amazing. Same voltage (18V9Ah) and amazing. The difference in max volume is crazy.
So for me TPA3116 is not the best.... It's cheap but not enough power.
TAS5611 has 12.6A peak current. so more than twice as powerfull.
I have 100mOhm shunt.
Later I will do a mesurement.
What I forgot to say is that first time I hade red dual TPA3116 (2x 100W) board fopr driving two JBL Control 5 with same problem.
Just amazing. Same voltage (18V9Ah) and amazing. The difference in max volume is crazy.
So for me TPA3116 is not the best.... It's cheap but not enough power.
TAS5611 has 12.6A peak current. so more than twice as powerfull.
I have 100mOhm shunt.
Later I will do a mesurement.
What I forgot to say is that first time I hade red dual TPA3116 (2x 100W) board fopr driving two JBL Control 5 with same problem.
Last edited:
Sorry for the typo, The R1 resistor is 47 K Ohm, so maybe it's a little to much if the reccomended one is only 5k
Well, thanks for the suggestion, if my pcb won't work out I could try a different and better one, but I need at least a volume pot.
I listened to your MP4 again on bigger speakers and as far as I can hear there is considerable distortion in the bass even before the amplifier cuts-off.
I have the same type of music I use for test of amplifiers because it has a very dominant and deep bass. When I play such music, my bass drivers work hard and in particular I pull a lot of current in the amplifiers. It is a good test of the potential of both the supply and amplifier. More current is demanded than I imagined.
I will propose, if you have an oscilloscope, to put it across the amplifier supply terminals while you increase the volume until the amplifier cuts-off. With the 19V adapter for a start. The TPA3116 should have a current limit higher than the power adapter. How much voltage swing? Do we reach the limit of the supply?
It is only about if the LOW-limit is passed. During the start-up, the input-pin(s) is sampled and that determines the gain. As you have much less noise, I guess you are down to 20dB gain.
If you can easily put around 5K on top - try it.
Hi,
so I added a little more than 5k resistance to R1 and as you said before 1k resistor on input to the Right and 1k to the Left channel shorted to ground. I can say that the hissing is almost as before this 2 mods.
Is there anything else I should try? The 4k7 (it's ok if I use only 4k?) resistors on pins 4 5 10 11 capacitors? In this case any ground is ok?
Thanks!
4K is fine. There is only one ground but you can use that ground where you find it nearby the capacitors.
Is the shunt for the whole amplifier or for one channel only.
I see levels in the order of 4A (100mOhm shunt, 100mV per division on the scope). It's close to the current limit of the 19V laptop supply.
its the hole amplifier. The amplifier has 2x3300uF and stock 2x2200uF capacitor.
Powersupply is Lithiumbattery. So no ground problems. Battery can handle more than 18A permanent and BMS has 50A peak shutdown. So no problem with that part.
I thing some times the amp gets higher peaks but dificult do save them^^
Powersupply is Lithiumbattery. So no ground problems. Battery can handle more than 18A permanent and BMS has 50A peak shutdown. So no problem with that part.
I thing some times the amp gets higher peaks but dificult do save them^^
Hi Hans,
I think I know what your 3116 Problem is. You take a passive x-over and bi-amp it.
Because of the high pass beeing separated by a capacitor, the amp has no load in the lower frequncy region. The 3116 does not like it to be driven into a virtually open output. To verify this, just try putting a (rough guess) 12-20 ohm resistor at the output of the amp driving the tweeter, then connect the x-over and tweeter.
To get around this problem, use an active cross over and remove the x-over.
Lege einfach einen 12-20 Ohm Widerstand über + und - des Hochtonverstärkers, damit der Chip einen ohmsche Last hat, für die niedrigen Frequenzen. Es reicht für einen Versuch ein kleiner Widerstand, mehr, als das der Widerstand nach kurzer Zeit durchbrennt, kann nicht passieren.
I think I know what your 3116 Problem is. You take a passive x-over and bi-amp it.
Because of the high pass beeing separated by a capacitor, the amp has no load in the lower frequncy region. The 3116 does not like it to be driven into a virtually open output. To verify this, just try putting a (rough guess) 12-20 ohm resistor at the output of the amp driving the tweeter, then connect the x-over and tweeter.
To get around this problem, use an active cross over and remove the x-over.
Lege einfach einen 12-20 Ohm Widerstand über + und - des Hochtonverstärkers, damit der Chip einen ohmsche Last hat, für die niedrigen Frequenzen. Es reicht für einen Versuch ein kleiner Widerstand, mehr, als das der Widerstand nach kurzer Zeit durchbrennt, kann nicht passieren.
Hi Turbo
Thanks for you reply.
I have Crossover with coil and capacitor for medium and tweeter. So is this the problem? Four pieces are inside the woofer.
Same problem also with TOA SL-152, JBL Control 5 and Cabasse ....
Thanks for the german part
Thanks for you reply.
I have Crossover with coil and capacitor for medium and tweeter. So is this the problem? Four pieces are inside the woofer.
Same problem also with TOA SL-152, JBL Control 5 and Cabasse ....
Thanks for the german part
Attachments
I planed to use these D-amps for a bi-amped "active-passive frequency divided" box too and fell over this problem. I use DSP´s now, as they are cheaper than any decend passive x-over components anyway.
As far as I have experienced, it is no good idea to put a capacitor in line with the load of these chip amps. Up to the cross over frequency of capacitor/ load resistor combination the amp sees an open output (the load decreses with frequency). Which might /will trigger the overload circuit of the chip, if low frequency´s are played or even destroy it.
It is usualy a special feature of an amp, if it can drive a pure capacitive load.
A possible solution:
For example, if a 30 ohm load would satisfy the 3116´s amps output, in extreme this would be about 7 watt´s "burned" in a parallel resistor, which would make a 5 watt item enough for music, even played very loud.
Also, praktisch einen Kondensator in Reihe mit einem Lastwiderstand an den Ausgang des Verstärkers klemmen, bedeutet das mit sinkender Frequenz der Widerstand verschwindet. Der 3116 mag einen offenen Ausgang gar nicht, manche brennen dann sofort durch. Früher gab es jede Menge Transistorverstärker, die das auch machten, war natürlich blöde. Die Frage ist jetzt für Dich, reicht vielleicht ein 30 Ohm Widerstand, der im Bass nur ganz wenig Strom in Wärme umwandelt oder sogar ein noch höherer Wert, damit die Ausgänge zufrieden sind. Schau, bei 3 Ohm 70 Watt bedeutet bei 30 Ohm 7 Watt Leistung. Sollten also 30 Ohm reichen, wäre ein 5 Watt Widerstand sicherlich im Betrieb völlig ausreichend. Ich bekomme böse Schimpfe wenn ich nur Deutsch schreibe, zu recht in einem englischen Forum, finde ich.
As far as I have experienced, it is no good idea to put a capacitor in line with the load of these chip amps. Up to the cross over frequency of capacitor/ load resistor combination the amp sees an open output (the load decreses with frequency). Which might /will trigger the overload circuit of the chip, if low frequency´s are played or even destroy it.
It is usualy a special feature of an amp, if it can drive a pure capacitive load.
A possible solution:
For example, if a 30 ohm load would satisfy the 3116´s amps output, in extreme this would be about 7 watt´s "burned" in a parallel resistor, which would make a 5 watt item enough for music, even played very loud.
Also, praktisch einen Kondensator in Reihe mit einem Lastwiderstand an den Ausgang des Verstärkers klemmen, bedeutet das mit sinkender Frequenz der Widerstand verschwindet. Der 3116 mag einen offenen Ausgang gar nicht, manche brennen dann sofort durch. Früher gab es jede Menge Transistorverstärker, die das auch machten, war natürlich blöde. Die Frage ist jetzt für Dich, reicht vielleicht ein 30 Ohm Widerstand, der im Bass nur ganz wenig Strom in Wärme umwandelt oder sogar ein noch höherer Wert, damit die Ausgänge zufrieden sind. Schau, bei 3 Ohm 70 Watt bedeutet bei 30 Ohm 7 Watt Leistung. Sollten also 30 Ohm reichen, wäre ein 5 Watt Widerstand sicherlich im Betrieb völlig ausreichend. Ich bekomme böse Schimpfe wenn ich nur Deutsch schreibe, zu recht in einem englischen Forum, finde ich.