Of course you are going to get a huge peak like that in an unloaded class D amplifier. The output filters have a very high Q alignment when unloaded. It is this peaking that TI caution you against when configuring the amplifier for post filter feedback.
With a load attached this peaking disappears but without a load it can cause gross instability if the phase margins are exceeded.
Try measuring the same thing but with the amplifier loaded. Either resistively or inductively via a typical tweeter impedance.
Of course if you load the amplifier with a woofer directly then the seen impedance will be very high at the resonant frequency of the output filters, but in this case it'll most likely be in an active system with a low pass of low frequency, limiting the bandwidth of the signal reaching the amplifier.
With a load attached this peaking disappears but without a load it can cause gross instability if the phase margins are exceeded.
Try measuring the same thing but with the amplifier loaded. Either resistively or inductively via a typical tweeter impedance.
Of course if you load the amplifier with a woofer directly then the seen impedance will be very high at the resonant frequency of the output filters, but in this case it'll most likely be in an active system with a low pass of low frequency, limiting the bandwidth of the signal reaching the amplifier.
I totally agree with you in each point. There is no need to convince me that damping comes in with a real load - this is a well known fact to me. I use these guys for driving hi-inductive guitar-speakers and the signal from the pre-amp is well bandwidth limited. All in all this is more of a theoretical issue, relevant only for the ultra-wide-bandwidth fraction of DiyAudio
Congratulations!!
Very nice job.
Did you chose the 2.1 config because of your mobile setup?
Is there any 2.0 or mono pbtl product available sporting this chip?
I would like to hear how much of an improvement it is compared to a TAS5630 in pbtl.
Hi Bühgemeiste,
I got a board with 2.0 configuration "up and running". I call this amp HCSA. Last weekends I found some time for the first serious measurments and was quiete happy about what I saw. You can use this Amp with or without DSP. My THD+N measurments follows the datasheet of TPA3255, with active DSP the THD+N is between 0.007% - 0.01% in most of cases. Without DSP it's a bit better.
I will post all measurments till the end of week.
Link to my thread: http://www.diyaudio.com/forums/class-d/298393-my-new-tpa325x-dsp-bluetooth-x-amplifier-made-mobile-application-2.html
Regards
Bernhard
Yes, the 2.1 config is for the bike setup, giving a little more "ompf" for the bass. The board can also be configured as 1.0 PBTL, 2.0 BTL, 2.1 BTL+2SE or 4.0 SE.
If anyone is looking for a Christmas present:
http://www.diyaudio.com/forums/swap...-0-btl-amplifier-modell-tiny.html#post4893402
http://www.diyaudio.com/forums/swap...-0-btl-amplifier-modell-tiny.html#post4893402
I have read through your post doctormord.
And you have done some great work.
I was at you webpage where i see that you have been working on a module using SRC4392,ADAU1701 and PCM5102A.
Now i work with the ADAU as well. But why do you not skip the ADAU1701 and replace the DAC by PCM5242? It has a build in MiniDSP which also has biquad filters taps. Enough for doing a crossover in the digital domain.
BR
Sonny
And you have done some great work.
I was at you webpage where i see that you have been working on a module using SRC4392,ADAU1701 and PCM5102A.
Now i work with the ADAU as well. But why do you not skip the ADAU1701 and replace the DAC by PCM5242? It has a build in MiniDSP which also has biquad filters taps. Enough for doing a crossover in the digital domain.
BR
Sonny
When i made the ADAU1701 boards, the PCM5242 wasn't an option. (It's quite new - 2014)
Beside this, getting the TI miniDSP software isn't as easy as getting SigmaStudio from AD. (If i remember correctly, you'd have to apply for a license.)
Did i ever showed pictures of the finished board?
Beside this, getting the TI miniDSP software isn't as easy as getting SigmaStudio from AD. (If i remember correctly, you'd have to apply for a license.)
Did i ever showed pictures of the finished board?
Attachments
Yo ho,ho!
Hi Doc;
I have been monitoring the exchanges regarding this chip
I have been deliberating the EVM from Mouser over this last week
What Power Supply have you found to be the best match with the 3255?
I may forgo the EVM and ask Santa to pm you for your payment details - instead of fannying about trying to put the EVM together with Inputs/Outputs etc, etc
I'm in the UK
Thanks
Owen
Hi Doc;
I have been monitoring the exchanges regarding this chip
I have been deliberating the EVM from Mouser over this last week
What Power Supply have you found to be the best match with the 3255?
I may forgo the EVM and ask Santa to pm you for your payment details - instead of fannying about trying to put the EVM together with Inputs/Outputs etc, etc
I'm in the UK
Thanks
Owen
Hi, all of you guys.
I've been staring at this chip for a while and have now a handful of them sitting on my desk.
I first thought that I'd layout the PCB myself, but as I have not any SW ready set-up to do it -and using ours at work, out of working hours, would at leat be tedious to me- I'm really happy that I found this (really exciting) website and thread.
I'm not yet really into it, but hope soon to be able to discuss with you in depth about any detail of it, if my humble 10 cents could help...
Regards,
Kal
I've been staring at this chip for a while and have now a handful of them sitting on my desk.
I first thought that I'd layout the PCB myself, but as I have not any SW ready set-up to do it -and using ours at work, out of working hours, would at leat be tedious to me- I'm really happy that I found this (really exciting) website and thread.
I'm not yet really into it, but hope soon to be able to discuss with you in depth about any detail of it, if my humble 10 cents could help...
Regards,
Kal