Don't forget the EVM Manual! Clip LED circuit is on page 18 lower right corner. It's also brought out onto the daughter board pin leads.
http://www.ti.com/lit/ug/slou441/slou441.pdf (D2, in specific)
http://www.ti.com/lit/ug/slou441/slou441.pdf (D2, in specific)
Yes, I'm grateful. I honestly studied those and the 3255 datasheet to the extent I could, though I wasn't particularly clear about that, but it seems to me the fault and clip pins from the 3255 chip are connected directly to the big J28 interface connector (just left of center on page 17).
So far I know about three alternatives:
1) Light pipes to the existing on board leds, as helpfully suggested above,
2) Take the active low fault signals from J28 and copy the 2 identical mosfet circuits as used on the EVM to drive 2 more leds on the case panel.
3) Live without external fault indicators.
My hope is that there might be a simpler circuit than 2. But I'm not experienced or bright enough to say what it might look like. If there is a simpler circuit, that's the one I really want to use. I am encouraged to believe in its existence by the fact that TI bothered to export the clip and fault pins to the J28 interface connector, though it's possible it wasn't designed to cater for my personal level of expertise.
I am basically hoping someone comes along and says 'obviously just attach an led here...' or some such. A shot in the dark.
So far I know about three alternatives:
1) Light pipes to the existing on board leds, as helpfully suggested above,
2) Take the active low fault signals from J28 and copy the 2 identical mosfet circuits as used on the EVM to drive 2 more leds on the case panel.
3) Live without external fault indicators.
My hope is that there might be a simpler circuit than 2. But I'm not experienced or bright enough to say what it might look like. If there is a simpler circuit, that's the one I really want to use. I am encouraged to believe in its existence by the fact that TI bothered to export the clip and fault pins to the J28 interface connector, though it's possible it wasn't designed to cater for my personal level of expertise.
I am basically hoping someone comes along and says 'obviously just attach an led here...' or some such. A shot in the dark.
You seem to be kind of suggesting option 2, duplicate the EVM circuits, which I have thought about.
But then I need to decide how badly I want these leds. Probably not badly enough to justify building and installing that particular circuit twice. Though it might be fun it would also delay completion (again).
You certainly helped to answer my woolly question, so thanks for your responses.
But then I need to decide how badly I want these leds. Probably not badly enough to justify building and installing that particular circuit twice. Though it might be fun it would also delay completion (again).
You certainly helped to answer my woolly question, so thanks for your responses.
Hi, I saw the TPA3255EVM module in the TI when searching for ICs to power my 2 * 100W @ 8 ohm speakers, which I think is the best for it with a good price (correct me if I 'm wrong and pls tell me better options). And the fact that I can change the jumpers to change the modes is excellent for the long run if I want in the future to couple two of these in mono but as I am a student and money is not that much I was thinking on diy through the schematics they offer to lower the price.
I know there is some similar Chinese ones, but they are locked to one time of mode 2.0 or 2.1 and it has cheaper components too.
Should I try or is it to much trouble for me to handle?
Maybe its more worth if i find this module at 50% like they did before, do you know when they do these promotions?
I know there is some similar Chinese ones, but they are locked to one time of mode 2.0 or 2.1 and it has cheaper components too.
Should I try or is it to much trouble for me to handle?
Maybe its more worth if i find this module at 50% like they did before, do you know when they do these promotions?
I have been enjoying this thread.
All I know about class D is the Crest/Peavey amplifier I use to drive my Edgar Seismics. I am frustrated by these because of their extreme gain. they are noisy - not absurdly but more than anyone would prefer. I do not liek the idea of a potentiometer set so low in its range. I use only one channel - i have two of the next to the lowest power PROLITE series - #3, I think. It doesn't sound bad I must say. I have been pleased enough with it but know that I can do better.
When I started reading these posts it was with the hope that this might make for a better replacement. The amps above this are First Watt J2 and SIT 1.
I am using a YAMAHA YDP2006 as the bass crossover. I have made many what I think are improvements to it. Eliminating all superfluous buffer stages and setting the gains to work best with my system. I know with the TI amp I will need to change the gain for the channel feeding the SEISMICS. The other channel is for a RYTHMIK below 50 hz. One of everything per channel.
I can either use the balanced output stage of the YDP or convert it to single ended. I have converted the input to single ended. Just to keep my obsessionist credentials in order i replaced all of the 5532s with LME49710s.
I think they sound more tonally sure of themselves than they did before.
So with all of that you know there will be questions which i hope one of you knowledgeable folks will be willing to give me some counsel.
I just got a module in today. TI did honor the4 discount on this and it took about a week to get here. I did have an existing account from them. I have got some power regulator EV boards in the past.
These SEISMICs are not as Edgar intended. I have optimized them for upper bass - 50 to 150 hz. The box behind the woofer has been greatly reduced in size so the woofers are under much greater control than they were in the original where they could flop around and when fed a boost signal they could really flop around.
The point being I do not need super low bass power. I would like to get as much control as possible. I am thinking about running the amps in mono but if told to consider using them as stereo (I still would only use one of the two channels since I prefer the amps close to the speaker boxes).
To begin:
Has anyone come up with a guideline as to when to use good smps or linear supplies? I am one who tends to always want linear. I look at what it would cost to make a good linear supply and i see I am getting close to the price of an appropriate MEANWELL - three hundred dollars. I know that the smps has a greater ability to deliver current quickly and this is very important. I gues I am nervous about the noise spec - 250 mV - this is at 200KHz and I have no idea how this compares to 1 mV at 60 - CRC modeled with PSU Designer so I know it is just a prediction. Anyone have an opinion on the best way to go for this frequency range? I feel pretty sure if my intention was to use it full range I would go with linear. For bass I could see smps having some advantages - not that I have any experience with this. Just conceptually it seems that way.
Any opinion on whether to go single ended or balanced input - of course, this is intertwined with the output configuration. I am not enamored of balanced - the rest of my system is single ended. Are the advantages of a balanced connection more important in the low or high frequencies? If for the low I would stick with balanced.
I know all of these questions do not have absolute answers. I am sorry i have gone on so long. What I would like is for others to say what they have done within their systems and why they did what they did. I think it would be interesting and useful.
All I know about class D is the Crest/Peavey amplifier I use to drive my Edgar Seismics. I am frustrated by these because of their extreme gain. they are noisy - not absurdly but more than anyone would prefer. I do not liek the idea of a potentiometer set so low in its range. I use only one channel - i have two of the next to the lowest power PROLITE series - #3, I think. It doesn't sound bad I must say. I have been pleased enough with it but know that I can do better.
When I started reading these posts it was with the hope that this might make for a better replacement. The amps above this are First Watt J2 and SIT 1.
I am using a YAMAHA YDP2006 as the bass crossover. I have made many what I think are improvements to it. Eliminating all superfluous buffer stages and setting the gains to work best with my system. I know with the TI amp I will need to change the gain for the channel feeding the SEISMICS. The other channel is for a RYTHMIK below 50 hz. One of everything per channel.
I can either use the balanced output stage of the YDP or convert it to single ended. I have converted the input to single ended. Just to keep my obsessionist credentials in order i replaced all of the 5532s with LME49710s.
I think they sound more tonally sure of themselves than they did before.
So with all of that you know there will be questions which i hope one of you knowledgeable folks will be willing to give me some counsel.
I just got a module in today. TI did honor the4 discount on this and it took about a week to get here. I did have an existing account from them. I have got some power regulator EV boards in the past.
These SEISMICs are not as Edgar intended. I have optimized them for upper bass - 50 to 150 hz. The box behind the woofer has been greatly reduced in size so the woofers are under much greater control than they were in the original where they could flop around and when fed a boost signal they could really flop around.
The point being I do not need super low bass power. I would like to get as much control as possible. I am thinking about running the amps in mono but if told to consider using them as stereo (I still would only use one of the two channels since I prefer the amps close to the speaker boxes).
To begin:
Has anyone come up with a guideline as to when to use good smps or linear supplies? I am one who tends to always want linear. I look at what it would cost to make a good linear supply and i see I am getting close to the price of an appropriate MEANWELL - three hundred dollars. I know that the smps has a greater ability to deliver current quickly and this is very important. I gues I am nervous about the noise spec - 250 mV - this is at 200KHz and I have no idea how this compares to 1 mV at 60 - CRC modeled with PSU Designer so I know it is just a prediction. Anyone have an opinion on the best way to go for this frequency range? I feel pretty sure if my intention was to use it full range I would go with linear. For bass I could see smps having some advantages - not that I have any experience with this. Just conceptually it seems that way.
Any opinion on whether to go single ended or balanced input - of course, this is intertwined with the output configuration. I am not enamored of balanced - the rest of my system is single ended. Are the advantages of a balanced connection more important in the low or high frequencies? If for the low I would stick with balanced.
I know all of these questions do not have absolute answers. I am sorry i have gone on so long. What I would like is for others to say what they have done within their systems and why they did what they did. I think it would be interesting and useful.
Thanks for the fast response 
About having to desolder, you're talking about the Nobsound china's one right?
Yeah I tested it right now and it still works which is awesome. When I saw that the thread "TI Class D EVM Board 50% Promotion"" had already a year old I didn't even tried to use it because I though it didn't work.
I'm liking more and more that texas company
About having to desolder, you're talking about the Nobsound china's one right?
Yeah I tested it right now and it still works which is awesome. When I saw that the thread "TI Class D EVM Board 50% Promotion"" had already a year old I didn't even tried to use it because I though it didn't work.
I'm liking more and more that texas company
Hi Rick, as far as smps goes these newer TPA amps are immune to a good smps. I am plenty picky about power supplies and just about every other diy project I own uses linear, except for the TPA. Maybe a linear sounds better, but these amps sound so good as they are I just don't care. The meanwells are popular but I prefer the connexelectronic smps's. They are well made and compact with no sonic drawbacks IMO. In fact I believe doctormord has said that the TPA amps require or perform best with a certain amount of mV. So lowest noise is not necessarily of benefit.
It hums when you're rotating it, or standing still?
If its the former, it can be your potentiometers, there is two ways you can solve it: changing them or put a capacitor at the output, check this: https://www.reddit.com/r/AskElectronics/comments/3al9wo/how_to_remove_noise_from_potentiometer/
Being the latter check: https://www.youtube.com/watch?v=-DwXpAlIYnA
Look I'm no pro, just trying to help so be careful.
If its the former, it can be your potentiometers, there is two ways you can solve it: changing them or put a capacitor at the output, check this: https://www.reddit.com/r/AskElectronics/comments/3al9wo/how_to_remove_noise_from_potentiometer/
Being the latter check: https://www.youtube.com/watch?v=-DwXpAlIYnA
Look I'm no pro, just trying to help so be careful.
Thanks for the fast response
About having to desolder, you're talking about the Nobsound china's one right?
Yeah I tested it right now and it still works which is awesome. When I saw that the thread "TI Class D EVM Board 50% Promotion"" had already a year old I didn't even tried to use it because I though it didn't work.
I'm liking more and more that texas company
Nope, the TI EVM's have a soldered-on tab that shorts out the output coupling capacitors (since in bridged configuration both speaker terminals are floating to the same DC value). This would be a VERY BAD THING in single ended, so you need to desolder the tabs to ensure the output is AC coupled.
I know nothing about the Nobsound parts. At $75 USD the TI EVM is very very hard to beat.
That way I dont think that the problem is the potentiometer.
Btw before I misslinked the video try apply a ground loop isolator check this to see if it can do any good: https://www.youtube.com/watch?v=qNQX8jyxRrs
Glad you told me that. From the schematics I though they've used jumpers which would be a better idea. But still nothing that gets trouble