Sonic Impact 5066 Parts List & Modifications

[Panelhead]

The film caps strapped across the speaker terminals in my SI are 0.15 ufd, not the 0.01 listed in the Tripath info. I think they made changes in the front, the circuit looks a lot simpler than the Tripath evaluation board.
The C3/C4 are the input caps. Polarity can be verified, the side that connects to the 20K resistor is the high voltage side.
I wonder is a high quality, 2.2 ufd is available. The ones listed in the Trpath data sheet are 2.2 ufd 10v ceramic X5R. The part number at DigiKey is PCC1868CT. These are multi layer type. They list them in larger values, but all look to be larger than the coupling caps used by SI. The one used by Tripath is larger than the SI.
Guess I need to make probes to fit my capacitance meter and check the value of the caps used by SI. I would be great to be able to extend the low end just by installing another cap here.

George

[motherone]

George,

Thanks for that piece of info! I totally missed that they spec'd it out on the datasheet. I think someone (Thorsten?) had posted in another thread that multilayer caps are terrible in signal chain due to smearing. I obviously have no idea on this, but it's food for thought.

I also agree that C3/4 look much too small to be 2.2uF. If you have a capacitence meter and can measure it, I'd be much obliged.

I'm hoping someone else can take a look at their boards and verify that my parts list above is correct. I'm guessing that if someone who has more experience at this type of stuff than I do can confirm it, it'll open modding these little guys up to even more people.

The one thing I still can find are the Schottky diodes mentioned on the datasheet. All the parts appear to be capacitors, resistors or coils, with no diodes in sight. Does anyone have any insight on this?

Thanks for your help!

[jkeny]

Another area that impacts sound significantly is the inductors. What is the make/quality of these on the SI? Keep up the good work. I have two SI's on order and hope to see them in the next couple of weeks (by which time this thread will have become an opus)

[Panelhead]

The one thing I still can find are the Schottky diodes mentioned on the datasheet. All the parts appear to be capacitors, resistors or coils, with no diodes in sight. Does anyone have any insight on this?

Thanks for your help!

The clamping diodes are there. They are the large square blsck surface mount components. I put mine with me back into the new case, but they sais BEC or something. I do not think these are high quality. This may be a good swap also.
When I return home later this week I will try to measure C3/C4.

George

[motherone]

Originally posted by jkeny
Another area that impacts sound significantly is the inductors. What is the make/quality of these on the SI? Keep up the good work. I have two SI's on order and hope to see them in the next couple of weeks (by which time this thread will have become an opus)

Someone had posted the inductor manufacturer, but I can't seem to find that thread. I also can't remember if it was posted on here, or one of the other forums. This would be good info, and I'd be more than happy to purchase some inductors to swap out and see how ti goes and return the results to the group, IF someone can suggest a suitable (better!) replacement.

Originally posted by Panelhead


The clamping diodes are there. They are the large square blsck surface mount components. I put mine with me back into the new case, but they sais BEC or something. I do not think these are high quality. This may be a good swap also.
When I return home later this week I will try to measure C3/C4.

George

George,

Which ones? There should be 4 diodes on board, and I can't find anything close to the chip that would indicate they're on here. The only black chips on my board are for the resistors (R01, R1, R2 and R4 on top, R1, R3, R5 R8 and R9 on bottom). There's also what I assume to be two inductors on the board, labelled L1 (top) and L2 (bottom). Everything else is labelled with a "C" for surface mount parts, so I've assumed those to be capacitors. The remaining parts are the LED, the 330uF electrolytic cap, and the 4 inductors.

Thanks,

Mike

[motherone]

Here's a quick list of possible inductor replacements for L1-L4. I'm not sure if someone out there knows more about what type of inductor would offer improved performance over the existing ones, but I thought I'd take a shot and post the candidates that I've found so far:

JW Miller Magnetics:
--------------------------

Part# 6000-100k, 10uH @ 3.4 Amps, DCR of 0.045, SRF of 14Mhz ($1.51 each)
http://www.digikey.com/scripts/DkSea...803&Row=323313


Toko America:
------------------

Part# 824MY-100K, 10uH @ 2.2 Amps, DCR of 0.052, SRF of 44Mhz ($5.57 each)
http://www.digikey.com/scripts/DkSea...455776&Site=US

[Pano]

Hi Guys,
Mike, you and George seem to be becoming real experts on these chips.

So of course I wanted to throw in my 2cents worth and also ask a couple of questions.

The caps called C2 /C4 are indeed the DC blockng input caps, i found them while poking around the board today.
They are also the reason for the bass roll off. There is no low end loss in the signal path until it gets to the DC side of these caps. Then the attenuation is the same as I measured and graphed (see my site).

I don't know what the 2 SMDs labled L1 & L2 do, but they are certainly in the signal path. No DC on either side, as far as I can tell. Can't tell if they are in series or paralelle with the signal path but they get the same signal as c2/C4. Why would there be an LC filter at the input?

The ouput inductors sing. If the amp is pushed hard (near max) the little chokes start to sing. You can quiet them down by placing a finger on them. Time for them to go.

Questions:
How are you ordering the Panasonic caps? I tried to follow the links posted but they had expired. I can find an in stock at Digikey, and minimums are 250~500.

Mike,
Did you get your Solen input caps in place yet? They should make a huge difference, if only in the bass. Have you also replaced the ouput filter?

I hope to make an seperate I/O board with better parts for the input and output filters. Also plan to bring the -3dB point of he output filter down an octave to about 35 kHz.

That's all for now.
Aloha
MM

My site: http://www.michael.mardis.com/sonic/

[motherone]

Originally posted by panomaniac
[B]Hi Guys,
The caps called C2 /C4 are indeed the DC blockng input caps, i found them while poking around the board today.
They are also the reason for the bass roll off. There is no low end loss in the signal path until it gets to the DC side of these caps. Then the attenuation is the same as I measured and graphed (see my site).

Thanks for confirming this. I'll be bypassing these guys soon.

Originally posted by panomaniac
[B]
I don't know what the 2 SMDs labled L1 & L2 do, but they are certainly in the signal path. No DC on either side, as far as I can tell. Can't tell if they are in series or paralelle with the signal path but they get the same signal as c2/C4. Why would there be an LC filter at the input?

I believe they are in parallel. The via right next to L1 is connected to ground, while the other pin looks like it's in parallel with the input.

Originally posted by panomaniac
[B]
The ouput inductors sing. If the amp is pushed hard (near max) the little chokes start to sing. You can quiet them down by placing a finger on them. Time for them to go.

Hmm. Maybe some physical dampening is in order? Either that, or some seriously better chokes should be put in. Perhaps the singing is caused by clipping?

Originally posted by panomaniac
[B]
Questions:
How are you ordering the Panasonic caps? I tried to follow the links posted but they had expired. I can find an in stock at Digikey, and minimums are 250~500.

Do a search for "Panasonic FM Capacitor" on digikey. Make sure you check the "In stock" button.

Here's the part numbers for a few of them:

Panasonic FM, 16v/680uF: P12377-ND
Panasonic FM, 16v/1000uF: P12366-ND
Panasonic FM, 16v/1200uF: P12367-ND
Panasonic FM, 16v/1500uF: P12368-ND

The 680uF should be a direct replacement for the existing electrolytic. It's the same size as the 680uF panasonic FC I'm using right now. The FM series is supposedly far superior to the FC series. Also, these are all available in quantities of 1, and the price ranges from $0.50 each to $0.76 each.

Originally posted by panomaniac
[B]
Mike,
Did you get your Solen input caps in place yet? They should make a huge difference, if only in the bass. Have you also replaced the ouput filter?

I'm waiting for my parts from digikey to arrive before starting on the next board. I'll probably throw this one in a wood enclosure for testing purposes and see how it goes. The parts from Digikey should arrive some time this week, so hopefully I'll be able to try it out this weekend.

Originally posted by panomaniac
[B]
I hope to make an seperate I/O board with better parts for the input and output filters. Also plan to bring the -3dB point of he output filter down an octave to about 35 kHz.

I plan on soldering the input caps directly to the back of the RCA jacks, and then hot-gluing or zip-tying them down in my enclosure.

As far as the output goes, I originally thought C21-24 were the output caps, but these are in fact the Schottky diodes mentioned on the datasheet. I whipped out the multimeter and confirmed it with it's diode testing function tonight.

R8b and C17 form the Right Channel Zobel, while R9b and C18 form the Left Channel Zobel. I think that the remaining caps on the bottom (C13, C14, C15, C16) and the two remaining ones on the top (C19 and C20) form the output network. I haven't had time to check the traces to figure out what's what. If you can measure these, we can probably just match them up to the datasheet.

Thanks for your input. I think we'll be getting pretty close to getting this board completely deciphered, which will make modding it even easier.

[Pano]

Originally posted by motherone
I believe they are in parallel. The via right next to L1 is connected to ground, while the other pin looks like it's in parallel with the input.

Hmmm.... I thought I saw signal on both sides of L1/L2. Will have to check again.

Maybe some physical dampening is in order? Either that, or some seriously better chokes should be put in. Perhaps the singing is caused by clipping?

It is, but some of the ringing just before clipping might be caused by the chokes. They need to go. Would we put inductors like this into our speaker projects? Let's hope not!

Here's the part numbers for a few of them:

Thanks a million.

I plan on soldering the input caps directly to the back of the RCA jacks, and then hot-gluing or zip-tying them down in my enclosure.

Sounds like a good plan. Simple and straight forward.

As far as the output goes, I originally thought C21-24 were the output caps, but these are in fact the Schottky diodes

Yes indeedy, they are. I took my switching waveforms from there.

R8b and C17 form the Right Channel Zobel, while R9b and C18 form the Left Channel Zobel. I think that the remaining caps<snip>

Good work! Nice to know where they are. We are getting close to figuring this thing out.

[Pano]

Here is the input schenatic for the Sonic. A bit different from the Tripath app notes.

This should help in bypassing any or all of it.