These boards have yet to be ordered or built, however, I have tried several boards - the the relatively-cheap $55 ones that are knock-offs of the TI evaluation boards on ebay and and Amazon (same ones apparently). The knockoff TI eval boards don't seem to work properly as the error/fault LEDs are lit upon startup (rather than lighting when there is a fault). I also tried to see what maximum output was and with a +52v 600w PSU feeding it, it could not put out anymore than 48Vpp - almost like it self-limited and resulting output started clipping. The knock-off boards also seem to be set up just for SE input and BTL output. It was my hope that we can have a general purpose board with all the features and options available and with a good layout and conveniences of locking connectors and thermal pad facing the mounting surface. The post filter feedback option is something we are looking into - it might not be possible to add without a major revision to the existing layout.
These are 2mm thick, 2oz boards with ENIG finish, hence the $32 price for just the bare PCB. If you want a professionally built PCB tested and ready to go, that will require substantial interest. Sorry about higher component prices in EU. You may look at Arrow, their shipping costs are very reasonable (not free anymore).
There is no way that you are going to beat the prices of the $55 Chinese-made TI knock-off boards with your DIY version (I think we all know that). This is an excercise to get the best performance and higher convenience and reliability. I would be using the cheap ones too if they did what I needed (I need 150w into 8ohms in BTL as an option on 1 board and I need a clean ~80w into 8ohms in SE output mode for 4 channels on another board).
These are 2mm thick, 2oz boards with ENIG finish, hence the $32 price for just the bare PCB. If you want a professionally built PCB tested and ready to go, that will require substantial interest. Sorry about higher component prices in EU. You may look at Arrow, their shipping costs are very reasonable (not free anymore).
There is no way that you are going to beat the prices of the $55 Chinese-made TI knock-off boards with your DIY version (I think we all know that). This is an excercise to get the best performance and higher convenience and reliability. I would be using the cheap ones too if they did what I needed (I need 150w into 8ohms in BTL as an option on 1 board and I need a clean ~80w into 8ohms in SE output mode for 4 channels on another board).
Last edited:
A "kit" with all parts could be possible but the amount of labor required to assemble the kits is not insignificant and like the professionally built one. depends on level of interest. If folks are interested in a kit with BOM, please indicated so. Note that the cost of a prepackaged BOM will be more than if you bought the parts yourself, so paying for convenience in the US. Outside of US, it may be less expensive if parts are hard to obtain. I have no idea on cost until I gather the parts and price them myself - have not gotten to that stage yet.
The TPA3255 is about $10ea, the Coilcraft SER inductors are about $5ea so we are already talking $30 for the major parts. We still have the DC-DC step up, the opamps, bulk caps, film caps, etc...
The TPA3255 is about $10ea, the Coilcraft SER inductors are about $5ea so we are already talking $30 for the major parts. We still have the DC-DC step up, the opamps, bulk caps, film caps, etc...
How do you swap tiny SMT opamps besides hot air pencil and desolder them? Also, if a great opamp is picked from the start, why would one want to swap them?
You can have your choice of opamps when you assemble it. Default is cost effective NE5532D, and if you want ultra performance, OPA1622 can be used at 5x the price.
OPA161x features (from manufacturer's website):
JPS64 is making progress. Note wide low impedance Vdd path from power input connector to the pins on the chip. Also relocating inductors for less crosstalk.
You can have your choice of opamps when you assemble it. Default is cost effective NE5532D, and if you want ultra performance, OPA1622 can be used at 5x the price.
OPA161x features (from manufacturer's website):
JPS64 is making progress. Note wide low impedance Vdd path from power input connector to the pins on the chip. Also relocating inductors for less crosstalk.
Attachments
Last edited:
I was referring to other threads where users of boards that have swappable opamps gave some interesting feedback on their subjective findings.
OPA1622 would be overkill imo.
AK4490 has NE5532 in it's official circuit, in which it produces <<-110dB distortion hands down, confirmed by my own measurements. The TPA3255 does about -100dB distortion under the best of conditions according to the Ti spec sheet. An opamp with better specs than the NE5532 will not lead to improved performance of the whole system.
AK4490 has NE5532 in it's official circuit, in which it produces <<-110dB distortion hands down, confirmed by my own measurements. The TPA3255 does about -100dB distortion under the best of conditions according to the Ti spec sheet. An opamp with better specs than the NE5532 will not lead to improved performance of the whole system.
Hi X,
Watching this with interest. I purchased a while ago 2 x TI TPA3255EVM boards when TI was offering the discount. If some of the design mod's (regarding components, not pcb via's, etc) you are planning can be implemented into the standard TI module, I hope you can share with the community without affecting your commercial endeavours.
Regards,
Myles
Watching this with interest. I purchased a while ago 2 x TI TPA3255EVM boards when TI was offering the discount. If some of the design mod's (regarding components, not pcb via's, etc) you are planning can be implemented into the standard TI module, I hope you can share with the community without affecting your commercial endeavours.
Regards,
Myles