|
|||||||
| Home | Forums | Rules | Articles | Store | Gallery | Blogs | Register | Donations | FAQ | Calendar | Search | Today's Posts | Mark Forums Read | Search |
| Class D Switching Power Amplifiers and Power D/A conversion |
|
Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving |
|
|
|
Thread Tools | Search this Thread |
|
|
#1 |
|
diyAudio Member
Join Date: Jan 2011
|
I purchased a couple of tiny boards based on the PAM8610 chip from Ebay. About $10. I want a tiny amp to feed my portable "Think-Geek" "crystal" speakers.
This chip feeds the class-D switched signal straight to the speakers, without any filtering. I added a few ferrite beads... Works OK, until you get a transient bass-drum. That sends the amp into fits, with a noise I will describe as a "racous burp." 8W+8W of sine wave at 30Hz is fine, the transient bass is what sends the PAM8610 into fits. Has anybody any idea where I should be looking to ameliorate the problem. I already have a huge cap on the power rail I suspect it is some form of bootstrapping with too lower a capacitor value... has anybody any suggestions?
|
|
|
|
#2 |
|
diyAudio Member
Join Date: Jan 2011
|
I went with my gut feeling and removed one of the bootstrapping capacitors to BSRN, BSRP, BSLN, BSLP which were easy to identify as C6, C7, C8 and C9.
It measured 570nF, pretty impressive for such a tiny cap, but less than the 1uF on the device data sheet. I dug some 1uF surface mount ceramics out of my toolkit and put them in parallel with C6, C7, C8 and C9.The $10 PAM8610 board now works really nicely. I used a plastic project box as a housing, but it fitted nicely into an Altoids tin. Actually, most people will not notice the original problem with bass transients. If you feed it with MP3, I suggest you may not even need the proper bootstrap caps. Anyway, as long as nobody laughs at my soldering technique at this micro-scale, here is a closeup of the modified C6 and C7. You will need to remove the electrolytic to get at C8 and C9. Oh - it works fine on 15V. I ran the voltage up to 17V on my own unit, just to make sure...
|
|
|
|
#3 |
|
diyAudio Member
|
Hi,
Can you comment on the sound quality of these modules? |
|
|
|
#4 |
|
diyAudio Member
Join Date: Jan 2011
|
If you desperately needed a low-power amplifier for, eg, headphones, and size was the only criteria, then maybe you might look at this board.
The biggest downside is that when it goes into clipping there are audible "burps" on each peak, rather than just distortion. Thus it really isn't much good for driving speakers, as it doesn't have enough power to spare to eliminate the possibility of clipping. |
|
|
|
#5 |
|
diyAudio Member
|
does't sound as if the sound quality has impressed you ? if you avoid clipping (and I like the built-in clipping indicator !) does it sound good ?
__________________
"The test of the machine is the satisfaction it gives you. There isn't any other test. If the machine produces tranquility it's right. If it disturbs you it's wrong until either the machine or your mind is changed." Robert M Pirsig. |
|
|
|
#6 |
|
diyAudio Member
Join Date: Apr 2011
|
Anyone had any experience of these? What's the sound like?
|
|
|
|
#7 |
|
diyAudio Member
Join Date: Apr 2011
Location: Coimbra
|
I'm bumping this and making the same question as fraction did. If you try and run sensite 8ohm speakers with this at moderate volume will it be interesting?
|
|
|
|
#8 |
|
diyAudio Member
Join Date: Feb 2010
Location: Tampa, Florida
|
trevmar, since you've done a little legwork experimenting with this board, I wonder if you can tell me a few other things. First, a lot of these class D chips seem to have selectable gain, and at leas on one (the Y148 Y148 Audio Amplifier Module - Free Shipping - DealExtreme ), I've noted that the board maker had the courtesy of lining up all the gain options, making a gain change as simple as moving a few 0 ohm resistors. Of course, as you already know, no board modifications are super simple on this size scale, so having the board at least set up for such mods is a huge help. Second, I need to be able to BTL tie the 2 outputs together to make a single MONO amp, better suited to a 4 ohm load. Most of these chips seem to make provision for this, but it has to be done at board design time, as simply tying the inputs 9and outputs) together is a "no no". Well "no no" or not, ant these prices I figure I can experiment some, and indeed with the Y148, though it is ill advised to do it this way, I did try just tying the outputs together along with the inputs, set the pot to full (to at least avoid any tracking error) and connected a 4 ohm speaker. It worked! in fact it worked so surprisingly well, with no apparent heating of the amp chip (at least to my calibrated finger ;-) So it makes me wonder if I could get away with doing the same with this PAM board. Ever try?
|
|
|
|
#9 |
|
diyAudio Member
Join Date: Jan 2011
|
Peter Pan, the Y148 is an infinitely better chip than the PAM chip. The board you linked to has an output filter, which is a major omission from the PAM. I discarded my PAM, I just couldn't get any power from it, and have been focusing on the TDA7492/TDA7498 modules from SURE Electronics and the L20D (for high-end systems). The SURE ST boards are pretty good, especially after modding them. I am just finishing off a webpage describing the SURE mods, and will summarize them in a thread here at the weekend
![]() . |
|
|
|
#10 |
|
diyAudio Member
Join Date: Feb 2010
Location: Tampa, Florida
|
Thanks trevmar. I figured as much, but for the size and price I can't resist trying to play with it. Yeah the Y148 has pretty amazing anti clipping feature too. Knowing how bad class D clipping can sound, and having connected it to an "unbridled" electric guitar with magnetic pickups, I was pretty taken back how well it handled at least a fair amount of transients. But as for that PAM board, there are at least a few apps I have in mind that could only be made possible by its small size, so I'll have to play with it a little. At least now i know about the bootstrap issue!
Having looked at the PAM chips specs, I see I can't do anything to increase its gain, but like I say... for its size, I can't resist checking it out. Only $10 to try, right? I HAVE ordered a few of those SURE electronics boards too. I'd layout and build my own, but my eyes and nerves are just not cut out for working this small anymore. I could lay out an cut a board, but stuff it? What a mess it would be! :-) Looking forward to your web posts! |
|
| Thread Tools | Search this Thread |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| 10w Xsoz | Mad_K | Pass Labs | 103 | 7th November 2004 05:10 PM |
| Best possible 10W from chip amp | Robert_66 | Chip Amps | 14 | 2nd November 2004 11:43 AM |
| JLH 10W Original | polsol | Solid State | 6 | 24th August 2004 07:43 AM |
| Heatsink for SOZ 10W | akira | Pass Labs | 4 | 21st May 2003 06:08 PM |
| JLH 10W integrated? | JP | Solid State | 1 | 25th September 2001 01:41 PM |
| New To Site? | Need Help? |