TPA3116D2 Amp

Jump to Latest
#2,141
Ians505 said:
So I finally got my two blue 2.0 ChengZhi boards in on Friday after about 6 weeks time. I spent the weekend putting them in aluminum cases (first time casing an amp board). Any suggested improvements are welcome.
I will be using them to actively power my monitors, one for the left and one for the right. I built the monitors some time ago but ran out of cash for amps (spent all my money on the best speakers and dsp I could afford). These 3116 amps are such a good price! They will be mounted to the back of the speaker enclosures close to the speakers. The signal and power cables will run up inside the speaker stands to feed the amps.
After I get to listen to these for a bit, I would like to try bypassing the LC filters and use ferrite beads as suggested in this thread. I'm not entirely sure how to accomplish this. Any suggestions are appreciated. I would also like to try larger caps if I can fit them.
Even though this thread is cluttered, there is still a lot of helpful info. Thanks to those who post useful information!
Click to expand...

Hi it would have been better to route the output cabling at one side of the case to the speaker connectors. Output cabling should be far away from input wiring/circuitry.
 
#2,142
This weekend, I assembled a little amp with a stock (no mods) blue YJ board.

One thing I've noticed that's kind of weird: with the amp connected to my speakers, powered on, but the input not connected, there is a slight but audible whitenoise/hiss coming through. This also happens if the inputs are connected, but the input source is powered off. My input source is a DAC. If the DAC is simply turned on (even with no actual audio signal), the hiss goes away.

Prior to this, I've been using an amp I assembled around the Sure tpa3110 board; it doesn't have this problem.

Also, and this may be related, with the tpa3116 amp: if I power it on or off without any input, it produces a loud POP from the speakers (I can see the woofers visibly move, kinda scary). But if the input (DAC) is connected and powered on, then the pop is minimized or non-existent when powering the amp on or off.

Any idea what's causing this? And what fixes might I look into?

Thanks!
 
#2,143
Just had a quick look at my blue board. I think YJ omitted resistors from the input to GND (ground reference resistors). If right the effect will be gone if you use let's say 100 kOhm resistors over your input connectors. So from input to GND, one 100 kOhm resistor at each channel.

So, again if i am not mistaking, the same effect will be seen when a volume potentiometer is used.
 
Last edited:
#2,145
The chip is irrelevant, it is about implementation of the simple GND reference resistors. Maybe the 3110 has those already.

Any capacitor coupled amplifier should have a resistor to reference inputs to GND (before the coupling caps). Without them the inputs are antennas. Secondly when you connect a device the input cap will be charged resulting in loud plops. The connected device will make the circuit and if there is some DC offset as is the case with these chips the caps will be charged. With a GND reference resistor the resistor will be the path for the cap to charge. As a bonus that charge pulse will also be amplified 😉

Just remember that 2 simple resistors can solve some possible nasties. Two ultrasmall 100 kOhm resistors can be soldered straight at the RCA connectors. The same will be valid when a potentiometer is used, those are resistors to GND. When using a volume potentiometer it is better to add a cap in front of that in each channel. I leave it to you to understand why that is.
 
Last edited:
#2,147
YBDZ 2.0 amps arrived

I finally got my two amps from Aliexpress after 8 weeks. It was a good price but that was too long. Hooked one of them up to my new dual chamber reflex speakers. They sound really good. Just what you expect - great bass and very clear highs and transparency. These will be part of my miniDSP bi-amp in a box setup. I like the screw terminal connectors for the speaker out and vcc in.
 

Attachments

  • image.jpg
    image.jpg
    382.8 KB · Views: 3,179
#2,148
I put 2 10000uf Rubycon caps in parallel with two 1000uf caps. Now it sounds amazing. I think 10000uf is not required for this amp but I can't find good quality higher capacity caps here so I will use them. Will it burn the chip or is there any side effect of that? Please advice.

3116.jpg

Thanks!
 
#2,150
Do you mind taking photos? I have the same altec lansing speakers so I might do what you did. I'd like to see how you did it. Thanks

bobistev said:
I have ended up getting the Feixiang 2.1 board for my Altec Lansing MX5021 office setup and have been using it for a couple of weeks now. To me the results are nothing short of amazing. These speakers were probably never designed to sound this good, the TPA just took them to a whole new level. The sound stage is broad and lifelike, while new sonic details emerge across the frequency range. The bristles shimmer with immediate presence and clarity as the brushes slide across the cymbal on Jobim's Getz/Gilberto interpretation of Desafinado. The sax passages ooze with warmth and depth, to my humble ears this sounds not too inferior to bulky 5 digit priced setups I have heard in hifi shop listening rooms.

I find the sub produces the best results with the crossover pot at around two thirds of the gauge (not sure why is this? - the pot should have a 80-500Hz range and the MX crossover point according to Altec docs should be at 150Hz).
For now I am using the naked board on a post of bubble wrap but I have spoken to the Aliexpress supplier about an enclosure, he contacted Feixiang factory for me and advised they should have an aluminium case ready for it in a couple of months. They also make a finished DAC (ebay.com/itm/360607831838) with the same front pot layout and dimensions so the amp case shouldn't look much different.
The amp is powered up by a run-of-the-mill laptop adapter rated at 18.5W & 4.9A.
Click to expand...
 
#2,154
Saturnus said:
Hopefully you mean parallel and not bridge.
Click to expand...

I mean bridging. Since one of the target application of this chip is for car amp use and bridging car amps to increase output is not uncommon. However, I could not find any information (may be I did not try hard enough) regarding bridging. I suspected that it may not be as simple as paralleling. Any comments or information are appreciated.

Regards,
 
#2,157
xrk971 said:
I don't think I have ever burned out a laptop SMPS ever - and I have had a lot over the years. They probably have built in self protection. I don't see any issue with using extra large power supply caps in the 10000 uF range - I would not worry about it.
Click to expand...

I don't think they will burn out. But someone told me that the power supply's over current protection will kick in when too much capacity is added. Those large caps require a very high charge current.
If this happens, the amp won't power up.
But i have never tested this myself.
 
#2,158
Lo_Tse said:
I mean bridging. Since one of the target application of this chip is for car amp use and bridging car amps to increase output is not uncommon. However, I could not find any information (may be I did not try hard enough) regarding bridging. I suspected that it may not be as simple as paralleling. Any comments or information are appreciated.

Regards,
Click to expand...

The TPA3116 is already internal bridged hence the designation BTL (bridge Tied Load). That means both positive and negative outputs are "hot".

You cannot bridge an amp that is already bridged. You can however parallel one or more outputs to be a PBTL (Paralleled Bridge Tied Load) amp.

Fortunately I think the TPA3116 is protected and will just shut down if you try, however I'm not certain and would not recommend trying.
 
#2,159
xrk971 said:
Isn't bridging the same as turning the amps into mono PBTL mode to get 100 watts/ch with 2 ohm drive capability? You are 'bridging' the left and right channels. That has been discussed earlier and there is even a mono PBTL board I think. Are are you talking about bridging multiple separate amps?
Click to expand...

Bridging means you take single ended output and tie the load across the load in either a full bridge or an H bridge instead of referencing to ground. TPA3116 however is already bridged so it cannot be bridged further. It can however be paralleled.
 
#2,160
In 3110 I added 10000 uF -20/+50% Roe large buffer elco's once to 1000 uF Panasonic FR's, maybe I should have taken the Panasonics out ? but result was not good, bass output was increased but seemed like it was all from reflex port or something, indirect and lacked definition, the Panasonic FR has strong but very defined bass, mids too and unfortunately highs too, a little too much highs I mean, allthought after some days that seemed to get better. The Roe elco's also added a "grid" "haze", don't know how to describe exactly, like a constant electronic haze/stress somewhere to the sound, seemed less open/relaxed with them connected.
 
You must log in or register to reply here.
Top Bottom