Finally got the TPA3250D2 card layout done. Decided to keep the Wiener theme going with the board naming, the card's now called the Wiener Pro. The layout is just as OCD as the TPA3118 Wiener card
For some reason the drill holes aren't showing up with this gerber viewer, but they're there.
Schematic is similar to what I posted earlier, but with the input diode clamps and gain selection jumpers hauled out. PCB size is 11x7cm, inductors are now ICE Components 1D17A instead of the original Coilcrafts. I've got a couple spots for through-hole 1/4W resistors on the input side, installing resistors will increase the gain. I'll configure the board for 20dB by default and throw in resistors for 26 and 32dB.
When I order boards for the next Wiener group buy in the coming weeks, I'll order boards/parts and build one up to try.
Hi gmarsh - This alone prompted me to join the forums
Cant wait for the Wiener Pro buy in thread to appear!
Has anyone on this thread actually built and listened to the 3251d2 in the PBTL configuration? I am looking at using this chip in that configuration as a bass/mid driver on a 4 ohm speaker where I am looking for a max output of about 100-120 watts with no more than 1% distortion at full output and preferably less distortion. . This will be single ended, only one speaker hanging on the end of this chip, there will be a separate amp for driving the tweeter and one of these chips in each enclosure.
DRMord,
do you see any problems with using the chip in this fashion, anything I should be looking out for or alternative ways to reach my goals for output and distortion level? I could just use a portion of the chip if that is a better solution but it just seemed like a practical solution to what I am looking for. Any advice is welcome.
do you see any problems with using the chip in this fashion, anything I should be looking out for or alternative ways to reach my goals for output and distortion level? I could just use a portion of the chip if that is a better solution but it just seemed like a practical solution to what I am looking for. Any advice is welcome.
So since you may be one of the very few people in the world to have actually heard an TPA3251d2 based amp playing - is it worth waiting for one of these over the currenty avaiable TPA3116 boards?
What power supply would you recommend for something like this?
Thank you. I can actually pick the voicecoil impedance so can optimize that for the amplifier and the output that I am looking at. I wouldn't go any lower than 4 ohms but could easily increase the impedance if that works out as a better match. I've found with my motor design that I can change the impedance between 4 and 8 ohms and still retain the same efficiency with the driver.
Dr Mord, quick question if you don't mind. Since you are one of the very few people to have had a chance to hear TPA3251 board play, would you consider waiting for one of these to be worth waiting over the currently available TPA3116 boards with a proven track record? Are they really better sounding?
Also, what power supply would you recommend for one of these?
Also, what power supply would you recommend for one of these?
Can't tell if they "sound" better. Compared to a TPA3132D2 (same as 3116) they tend to have a bit less noisefloor and just play louder.
The "sound" stays and fall with the layout. While this normally won't be heard at moderate levels, there will be differences at the top end of output power.
(As I never heard a Chinese made 3116/3118 board I can't compare to my 3132 board but I never had all the issues with noise/hiss etc. people here at Diyaudio complain about at some boards.)
Comparing a 50W 3116 vs a 175W 3251, it's just ~5dB more. So to have it twice aus loud, you would need 500W.
But while the 3116 is at 10% THD when doing 50W, the 3251 is below 1%.
Normal music levels at home are well below 1W, would say 50-100mW is a normal music listening level, so what counts then? Noise floor.
To reach the full power out of the 3251 you need a good supply, huge capacitor banks won't help when your supply is not at a low impedance. Also the layout needs to be low impedance to go the distance.
Conclusion:
The $38 board will do good, if they avoid the "typical" problems known from other (3116) boards. At higher levels THD might rise more compared to a design with "better components" I. the meaning of solid inductors, proper output filter caps, well designed analog frontend.
To which extend spending additional money makes "sense" is something that needs to get measured/evaluated.
If it works and satisfies you and the needs, all fine. But you'll always get what you pay for. Even if they can do a really good price, they can't source parts for nothing.
Maybe 5th element would measure on of these boards when ready. I'd donate some money.
I apologize for my bad english today.
Edit:
You can do a design for music and one for continuous sinewave output. While the first might/will fail on longer periods of pure sinewave output, the later will never fail on music and will have more thermal headroom for running it with bad/no ventilation, enclosures or higher ambient temp. But this usually costs money.
Power supplies are listed here in the thread, a MeanWell 36V 350W will do fine.
The "sound" stays and fall with the layout. While this normally won't be heard at moderate levels, there will be differences at the top end of output power.
(As I never heard a Chinese made 3116/3118 board I can't compare to my 3132 board but I never had all the issues with noise/hiss etc. people here at Diyaudio complain about at some boards.)
Comparing a 50W 3116 vs a 175W 3251, it's just ~5dB more. So to have it twice aus loud, you would need 500W.
But while the 3116 is at 10% THD when doing 50W, the 3251 is below 1%.
Normal music levels at home are well below 1W, would say 50-100mW is a normal music listening level, so what counts then? Noise floor.
To reach the full power out of the 3251 you need a good supply, huge capacitor banks won't help when your supply is not at a low impedance. Also the layout needs to be low impedance to go the distance.
Conclusion:
The $38 board will do good, if they avoid the "typical" problems known from other (3116) boards. At higher levels THD might rise more compared to a design with "better components" I. the meaning of solid inductors, proper output filter caps, well designed analog frontend.
To which extend spending additional money makes "sense" is something that needs to get measured/evaluated.
If it works and satisfies you and the needs, all fine. But you'll always get what you pay for. Even if they can do a really good price, they can't source parts for nothing.
Maybe 5th element would measure on of these boards when ready. I'd donate some money.
I apologize for my bad english today.
Edit:
You can do a design for music and one for continuous sinewave output. While the first might/will fail on longer periods of pure sinewave output, the later will never fail on music and will have more thermal headroom for running it with bad/no ventilation, enclosures or higher ambient temp. But this usually costs money.
Power supplies are listed here in the thread, a MeanWell 36V 350W will do fine.
Last edited:
tas5613 2.1 ampboard sellingprices start around $15, tas5630 2.0 ampboard sellingprices start around $27 at the moment in China, prices on YJ website are inflated already probably to make selling to resellers easier ? So pricedistance from ordering your own BOM is even bigger. What quality to expect is related to the $15 and $27 sellingprices, not the inflated YJ showprices $39 and $77