TPA3116D2 Amp

Jump to Latest
#5,301
erik777 said:
I re-looked at the chip schematic. I had the pins 4-5 & 10-11 reversed from amp in relation to the Edcor. So after i fixed that I knew I had resolved the problem. Immediately it sounded better which would be expected, but the volume loss was still the same 🙁 So I removed the center tab connection from Edcor coming from Amp. And waaalaaa, presto! that was the issue, but I dont know why. So now I was anxious to listen to a few tracks and test by switching from source to edcor then from source to amp. I can now say the detail is better, more accurate and its slighly more musical. Its not a night and day difference but its an upgrade that can be noticed. If i were to quantify in terms of %. Its from 20% on the low to as high as 30% depending on the type of music. The 30% would be with music that has alot different kinds of percussion. 20% would be in accuracy and musically. Perhaps it will even get better as it gets burned in.
I dont have a center tab connection from my laptop as its plastic to the Edcor. But I did have a center tab from the amp to the Edcor. I wonder if I had a cd player with a metal chassis that I could hook up to that, then to the Edcor input side, I wonder if that would cancel out the db loss, and improve the audio performance even more. I dunno----Thanks Chris and also to KJA2013 who was helpful.
Click to expand...

Thats great. Of course I have a vested interest in seeing the Ecdcors working as I want to get some impressions from people using them to determine if I make a purchase 😉 (investment greater for me as postage is very high to New Zealand)

I am no expert but I was thinking that one should always have balanced to balanced or unbalanced to unbalanced and not mix the 2? I think the way you had it effectively made the ratio 2:1 which gives the lowered voltage you observed.
 
Last edited:
#5,305
Hi. Newbie, etc.

I've always wanted to learn how to solder, and the Yuan Jing DIY kit seems like a good choice for a first project. Everything's there, including the YJ blue board, and if I find out I want to keep messing about, I could upgrade pretty much anything I wanted - I'd need to get someone to drill new holes in the enclosure, but that's it, really.

The only thing that strikes me as tough is soldering loose parts to the various small boards (just the power switch and pot). On the other hand, I figure that's probably a good second step after soldering wires to binding posts/ and RCAs to that odd three-pin board.

I will probably borrow tools, as I don't know if I'll continue. (If I do continue, next project will be the good ol' Adafruit SpokePOV.) It looks like I'll need a soldering iron with stand/sponge/soldersucker, DMM, and a good wirestripper. Anything else I should ask?

Thank you.
 
#5,307
erik777 said:
I re-looked at the chip schematic. I had the pins 4-5 & 10-11 reversed from amp in relation to the Edcor. So after i fixed that I knew I had resolved the problem. Immediately it sounded better which would be expected, but the volume loss was still the same 🙁 So I removed the center tab connection from Edcor coming from Amp. And waaalaaa, presto! that was the issue, but I dont know why. So now I was anxious to listen to a few tracks and test by switching from source to edcor then from source to amp. I can now say the detail is better, more accurate and its slighly more musical. Its not a night and day difference but its an upgrade that can be noticed. If i were to quantify in terms of %. Its from 20% on the low to as high as 30% depending on the type of music. The 30% would be with music that has alot different kinds of percussion. 20% would be in accuracy and musically. Perhaps it will even get better as it gets burned in. If you listen to heavy metal or similar, I dont think the Edcor would help much. But Jazz, classical, Orchestra, Blues, etc more complex types of music, it might be worth looking into. I dont have a center tab connection from my laptop as its plastic to the Edcor. But I did have a center tab from the amp to the Edcor. I wonder if I had a cd player with a metal chassis that I could hook up to that, then to the Edcor input side, I wonder if that would cancel out the db loss, and improve the audio performance even more. I dunno----Thanks Chris and also to KJA2013 who was helpful. Im listening to it right now as I edit this post, and it brings a pleasant smile. But now it makes me wonder more then ever. How will the Cinemags compare now ......hmmmmmm
Click to expand...

Have you removed the input coupling caps on the amp board? You described switching the transformers in and out, which suggests that the caps may still be in the circuit. If not, I'd strongly suggest that you do this. One of the main reasons to use input transformers is that they allow for the input coupling caps to be bypassed. The fair comparison is [amp with coupling caps and no input transformers] vs [amp without coupling caps with input transformers]. I think you'll be pleasantly surprised 🙂
 
#5,308
Last edited:
#5,309
sharpi31 said:
Have you removed the input coupling caps on the amp board? You described switching the transformers in and out, which suggests that the caps may still be in the circuit. If not, I'd strongly suggest that you do this. One of the main reasons to use input transformers is that they allow for the input coupling caps to be bypassed. The fair comparison is [amp with coupling caps and no input transformers] vs [amp without coupling caps with input transformers]. I think you'll be pleasantly surprised 🙂
Click to expand...


I do have the input caps still in place. I ran this by brother in law who is a EE. He tells me to leave it in, but ill run it by him again, and see what he thinks.
 
#5,310
erik777 said:
I do have the input caps still in place. I ran this by brother in law who is a EE. He tells me to leave it in, but ill run it by him again, and see what he thinks.
Click to expand...

Definitely bypass them! As Sharpi says, large part of the reason for using a transformer is to enable removal of the "evil" input cap! The transformer then performs the role of blocking any DC on the input.

Either remove or put a simple wire jumper across the input caps (depending how you wires the transformer) and make sure the -ve one is not earthed.
 
#5,311
ultracrepidarianist said:
Hi. Newbie, etc.

I've always wanted to learn how to solder, and the Yuan Jing DIY kit seems like a good choice for a first project. Everything's there, including the YJ blue board, and if I find out I want to keep messing about, I could upgrade pretty much anything I wanted - I'd need to get someone to drill new holes in the enclosure, but that's it, really.

The only thing that strikes me as tough is soldering loose parts to the various small boards (just the power switch and pot). On the other hand, I figure that's probably a good second step after soldering wires to binding posts/ and RCAs to that odd three-pin board.

I will probably borrow tools, as I don't know if I'll continue. (If I do continue, next project will be the good ol' Adafruit SpokePOV.) It looks like I'll need a soldering iron with stand/sponge/soldersucker, DMM, and a good wirestripper. Anything else I should ask?

Thank you.
Click to expand...

Go explore one of the build guides at a partner site Tubelab.com Getting Started | Tubelab
 
#5,312
undefinedza said:
*10uH.

Also note, the PBTL version won't give you much more power than BTL with a 4ohm load. As mentioned though, in PBTL mode you can drive a 2 ohm load and develop more power (thanks to higher current output ability of the chip in PBTL).
Click to expand...

2x8ohm wasn't a great success output power-wise: just 16watts based on my AC power meter with a 50Hz signal from my phone. There wasn't even that much movement when connected to the subwoofer output of my 5.1 amp at reasonable listening levels. Unfortunately I didn't have the power meter connected at the time.

Is there a difference between running this in PBTL with 2ohm load or in stereo with two 4 ohm loads? In a simple DC circuit there shouldn't be a difference (as far as I can tell) but perhaps this is different.

If I get nicer results from two shakers per amp I'll just get another amp and go with that.
 
#5,313
ChrisMmm said:
Definitely bypass them! As Sharpi says, large part of the reason for using a transformer is to enable removal of the "evil" input cap!
Click to expand...

Ill remove them today and post what i hear.The reason I went with Edcor 10k/10k input transformer and not 15k/15k. I was told its possible the 15k voltage could swing higher causing the amp to compress the signal, with little head room, resulting in the sound quality to be compromised. Maybe other experts can chime in on that one.
 
Last edited:
#5,314
jussik said:
2x8ohm wasn't a great success output power-wise: just 16watts based on my AC power meter with a 50Hz signal from my phone. There wasn't even that much movement when connected to the subwoofer output of my 5.1 amp at reasonable listening levels. Unfortunately I didn't have the power meter connected at the time.

Is there a difference between running this in PBTL with 2ohm load or in stereo with two 4 ohm loads? In a simple DC circuit there shouldn't be a difference (as far as I can tell) but perhaps this is different.

If I get nicer results from two shakers per amp I'll just get another amp and go with that.
Click to expand...


I have measured 77 watts into a 4ohm speaker with my amp at 50Hz. The heatsink is cooking by then but it runs fine and sounds great. Your PS voltage may be limiting you if only running at say 12volts into 8ohms.
 
#5,315
Figure13 datasheet shows you could expect a little more continuous. But that is powered from a PSU that doesn't limit amp, with TI EVM, no dead Jamicons which potentially are on your YJamp. Also maximum inputsignal I guess, which might not be reached in your situation??? and TI-EVM probably doesn't have inductors that become hot (energy) or other parts like that or possible powerleakage on switchingfrequency tpa3116 or smps psu, given a chance any classD will deliver max ampères to peaking filterfrequency too, also possibly drain PSU. All in all very reasonable result for that amp with that powersupply with possibly limited inputsignal.

2 ohms PBTL won't deliver more power with YJheatsink, instant max power is little above 100W a channel 4 ohms stereo for TI EVM
 
#5,316
xrk971 said:
I have measured 77 watts into a 4ohm speaker with my amp at 50Hz. The heatsink is cooking by then but it runs fine and sounds great. Your PS voltage may be limiting you if only running at say 12volts into 8ohms.
Click to expand...

I'm starting to believe that I will need something more powerful that a PBTL TPA3116 as I have just bought some speakers which will take 200Watts and are 4 Ohms. They appear to like being driven and my current chip amp is not ideal.

I am still awaiting the oscons for my sure board, so what voltage are you using to get 77 watts?

Anybody done similar mods on higher power boards If so which are good and how do they compare to the modded TPA3116 boards?
 
#5,317
toseland said:
I'm starting to believe that I will need something more powerful that a PBTL TPA3116 as I have just bought some speakers which will take 200Watts and are 4 Ohms. They appear to like being driven and my current chip amp is not ideal.

I am still awaiting the oscons for my sure board, so what voltage are you using to get 77 watts?

Anybody done similar mods on higher power boards If so which are good and how do they compare to the modded TPA3116 boards?
Click to expand...

Powerrating speakers isn't really relevant, do your ears need more SPL, I would try first.
 
#5,318
I don't know what impedance options they have for the transformers, but the output towards the amp would be well to be at least as low as the impedance at the gain you have, and up to 10x less. In fact this makes the amp "easier" to drive and acts somewhat like owning a buffer that has minimal loss instead of minimal gain.

Likewise, you want the side towards your volume/source to be high impedance than the volume/source, with the same general principle.

If none of that can be achieved, well, shoot, it's still probably better than any cheap input cap you could have.
 
#5,319
Destroyer OS. said:
I don't know what impedance options they have for the transformers, but the output towards the amp would be well to be at least as low as the impedance at the gain you have, and up to 10x less. In fact this makes the amp "easier" to drive and acts somewhat like owning a buffer that has minimal loss instead of minimal gain.

Likewise, you want the side towards your volume/source to be high impedance than the volume/source, with the same general principle.

If none of that can be achieved, well, shoot, it's still probably better than any cheap input cap you could have.
Click to expand...

Except that any cheap capacitor will most likely have a rather flat frequency respons compared to any transformer between any source and any amp. Resulting frequency respons of transformer isn't likely to be flat (automatically).
 
You must log in or register to reply here.
Top Bottom