My new working Class D TAS5630 DKD

Status
Not open for further replies.
Jump to Latest
#43
in single ended mode - yes.
In bridged mode (BTL, PBTL, AD or BD modulation) each input pin is a input for separate (paralleled) half bridge. So to get full power output, it is needed to feed a differential signal to the input pins.
 
#45
jash2008 said:
Firstly, I have shared the maximum that I could. Since I am selling these TAS5630 amps to many local customers. If you are intrested in PCB, I can sell it to you too🙂.

Comparing sound quality, I didnt found much difference in low frequencies but as stated in the datasheets, THD of 5630 is poorer than LM 3886. further a minute noise is also observed in 5630. Although if you compare the power output, 5630 is better in all respect.

I am having various drivers like 2 pairs of pioneer champion series 12" TS-W1208D2, NX audio 12" PS1260 and a 15" NX audio (dont remember model).
Click to expand...

yes ,
It will be of great help if you could give inverted images in pdf format,for photoresist developing.
thank you and regards
 
#49
Don't mean to hijack this thread... I built one of these amps based on the TI EVM PHD version (I used TI's Gerbers and had boards made). I built one up (not for the faint of heart - there are many, many components, most of them 0603). I made a laser cut solderpaste stencil and placed the parts using tweezers under a microscope. I have an SMD oven that I built for soldering (mandatory on this PCB, no way you can do a decent job by hand and I've been soldering SMD for 20 years). I hand wound the inductors (fun!) though subsequently I got some samples from a couple of companies in China. It fired up and worked perfectly the first time. This is one great amp. I don't have any audio test equipment (just Tek 2245A and Tek MSO2024 scopes). I did see a little bit of "furriness" on the output but I *almost* think this was a result of bad scope probing and not the amp. In a completely non-technical test I shorted the input to ground and could hear nothing (not a thing) from the speakers. With a Line 6 POD as the input and turned up all the way there was an "almost" barely audible bit of noise if you put your ear to the speaker, however, I know this comes from the Line 6 unit.

I used a switching power supply from a company that specializes in "audio SMPS" though I also got some quotes from Chinese companies for 1/4 the price that had the same specs (I never got one to test though). Anyway, this is a killer amp. I used it for music and as a guitar amp (using the Line 6 "POD" as a "front end") and was very impressed.

I still have a couple of the PCB's, however the manufacturer "took liberties" with the largest holes. On one set of boards he made the heatsink and mounting holes the wrong side and on another the holes for the big terminals too small for the spec'd terminals used to connect to the board. I'm willing to sell what I have left for short money if the hole thing doesn't matter to you (I soldered wires directly to the pads or left the terminals sticking up).

As for the heatsink, I got an old one and machined it down to fit in the space provided. I had in mind a design to use a PC heatsink and fan like Gaurav did but after probing around a bit on the EVM I think the output traces get too long between the IC and the caps. I guess you could machine the PC heatsink so the caps can be near the IC. The placement and way these are connected are critical to making this thing work correctly. If you roll your own follow what TI did to the letter. CRANKING this thing and keeping the "clipping" led on constant it didn't even get warm to the touch. With the heatsink and fan(!) shown here I'm sure there may never be a heat problem. I also agree with Marce; a two sided PCB is really mandatory for this IC. When you see the waveforms and current paths on the board, there is no way a single-sided PCB can perform to the spec. If you plan on building one of these, know ahead of time it is a huge undertaking (just ordering components can be an ordeal). Getting the inductors right is critical. Placing all the parts took me three hours (though soldering takes 4 minutes in the oven so that's a push).

Good luck!

Mike T.
 
#51
savu said:
Hello 81bas... could you give some examples of those critical places that you spoke about because i just taught it has to be linked to the ground and that's all 🙁

regards,

savu
Click to expand...

Hi! please see in http://focus.ti.com/lit/ds/symlink/tas5630.pdf at pages 25 and 26. These places/traces are marked with color arrows and bulbs...
As a small hint: the GND pins in this IC are NOT connected internally (in both DKD and PHD versions). So any smallest GND trace inductivity will produce huge voltage spikes between GND pins, which will definitely kill the IC...
 
Status
Not open for further replies.
Top Bottom