Finally got a couple hours to spend on the TPA3250D2 card.
Breakout's figured out, going for the same "stupidly low power supply Z" strategy as the Wiener card. With the heatsink no longer on the card, I've got all this extra room, trying to figure out what to do with it. Probably gonna shrink the card a little, right now it's 8x12cm.
Breakout's figured out, going for the same "stupidly low power supply Z" strategy as the Wiener card. With the heatsink no longer on the card, I've got all this extra room, trying to figure out what to do with it. Probably gonna shrink the card a little, right now it's 8x12cm.
There's actually no hiss at the 3251 as the AVDD is generated separately, so PSRR (normally) is way better on this rail until you set the system voltage so low that the regulators get into dropout.
I did a short test with the new heatsinks+fan for the small boards mounted on the (bigger) blue board. Without fan it seems that the amp will run "forever" with music program. (Set to light clipping, tested with Drum n Bass -> small crest factor)
Haven't tested for longer period of time as my load resistors desoldered.
Output power (both channels) was at an average of 100W with bursts at 250W.
(But: free air, 20C ambient, bigger pcb)
I did a short test with the new heatsinks+fan for the small boards mounted on the (bigger) blue board. Without fan it seems that the amp will run "forever" with music program. (Set to light clipping, tested with Drum n Bass -> small crest factor)
Haven't tested for longer period of time as my load resistors desoldered.
Output power (both channels) was at an average of 100W with bursts at 250W.
(But: free air, 20C ambient, bigger pcb)
Can't tell as i never heard a "chinese made" 3116 board. My 3132 board (which is a 3116 in qfn package) does not hiss.
Spec-wise, the 3251 is (a magnitude) better than the older generation 31XX ics.
For the AVDD, it's a total different story, as the 3151 has "virtually" no input stage. (Just 20dB inverting amplifier per channel, where the 3116 got a PGA and Diffamp)
So care must be taken in design of the "dicrete" analog frontend feeding the 3251.
PCBs are in the mail. (should arrive in 3 weeks i guess)
Spec-wise, the 3251 is (a magnitude) better than the older generation 31XX ics.
For the AVDD, it's a total different story, as the 3151 has "virtually" no input stage. (Just 20dB inverting amplifier per channel, where the 3116 got a PGA and Diffamp)
So care must be taken in design of the "dicrete" analog frontend feeding the 3251.
PCBs are in the mail. (should arrive in 3 weeks i guess)
Attachments
Last edited:
Powered my first one up.
Seems OK.
No smoke.
Only used 20V supply with 8R loads. Idle current 20mA.
Drove it slightly to about 1W out.
Have to get some cables for my Diff-SE amplifier so I can look at the output on the scope. Then I can push it more and (visually) see if I can get 75W/ch into 8R.
Then I will try for 150W/ch into 4R and see if 1OZ. copper can handle it.
Seems OK.
No smoke.
Only used 20V supply with 8R loads. Idle current 20mA.
Drove it slightly to about 1W out.
Have to get some cables for my Diff-SE amplifier so I can look at the output on the scope. Then I can push it more and (visually) see if I can get 75W/ch into 8R.
Then I will try for 150W/ch into 4R and see if 1OZ. copper can handle it.
Latest:
finished the BOM for the new boards, those Coilcraft inductors are a bit problematic to source, as they can only sourced from Coilcraft directly. We may have to overthink this.
Anyone know alternative high-current dual inductors which aren't made of carbonyl compound? (Which saturate soft)
Will setup one board within the next i guess, need to laser some stencils.
finished the BOM for the new boards, those Coilcraft inductors are a bit problematic to source, as they can only sourced from Coilcraft directly. We may have to overthink this.
Anyone know alternative high-current dual inductors which aren't made of carbonyl compound? (Which saturate soft)
Will setup one board within the next i guess, need to laser some stencils.
Hi Mike, good to know.
Today, i had the chance to try out our new laser engraver to do some stencils and case parts. (Went a bit unlucky due to to much power for the plastics i had but this gets fixed next)
Anyway, the first thing i had in mind was "The Rig" to fulfill the needs for a dampened rig to reduce distortion caused by vibration.. Hmm, joking? "The Rig" is made for bolting down the amp into a portable case and keep the flexing stress away from the board and ceramics. This needs a bit more research to find some good spring damper constants. The actual setup is not accounting the area focus.
Secondly i made a housing for "The red stinker". Top plate is made with a simple grill at the moment but it might be a good idea to have the fan right over the housing for an optimized air flow (in/out). The bottom plate is made so that the chimney effect is cooling the inductors and bulk caps.
Today, i had the chance to try out our new laser engraver to do some stencils and case parts. (Went a bit unlucky due to to much power for the plastics i had but this gets fixed next)
Anyway, the first thing i had in mind was "The Rig" to fulfill the needs for a dampened rig to reduce distortion caused by vibration..
Hmm, joking? "The Rig" is made for bolting down the amp into a portable case and keep the flexing stress away from the board and ceramics. This needs a bit more research to find some good spring damper constants. The actual setup is not accounting the area focus.Secondly i made a housing for "The red stinker". Top plate is made with a simple grill at the moment but it might be a good idea to have the fan right over the housing for an optimized air flow (in/out). The bottom plate is made so that the chimney effect is cooling the inductors and bulk caps.
Attachments
Last edited:
Looks like TI have just released a higher power version of the 3251D2. The odd thing is that it comes in exactly the same IC package. This makes me scratch my head as you can't really get the continuous full output power out of the 3251D2 without some crazy thermal solution.
The new chip has a much higher operating voltage so can swing more on its output. This is nice as it gives far more peak headroom and with music it's probably not an issue but still I worry about all that oomph going through such a tiny chip.
The new chip has a much higher operating voltage so can swing more on its output. This is nice as it gives far more peak headroom and with music it's probably not an issue but still I worry about all that oomph going through such a tiny chip.