Right now Viktor has a pile of assembled boards at his house 
Almost there.
Standard version will support both HP-10W and GW-1058's. The current cAMP audio amplifier is a TPA3118 amplifier configured for 4 ohm loads, with a couple audio stages in front of it providing a highpass filter and differential receiver.
Provided you put a 4 ohm load (or thereabouts) on it, it'll work fine. 6/8 ohm loads will cause a bit of HF peaking.
Almost there.Standard version will support both HP-10W and GW-1058's. The current cAMP audio amplifier is a TPA3118 amplifier configured for 4 ohm loads, with a couple audio stages in front of it providing a highpass filter and differential receiver.
Provided you put a 4 ohm load (or thereabouts) on it, it'll work fine. 6/8 ohm loads will cause a bit of HF peaking.
Right now Viktor has a pile of assembled boards at his house
Standard version will support both HP-10W and GW-1058's. The current cAMP audio amplifier is a TPA3118 amplifier configured for 4 ohm loads, with a couple audio stages in front of it providing a highpass filter and differential receiver.
Provided you put a 4 ohm load (or thereabouts) on it, it'll work fine. 6/8 ohm loads will cause a bit of HF peaking.
Can't wait. I've been buying pieces for my boominator month by month and have been waiting for this so I can put everything together. Keep us updated!
Update - here's our production card!
You may notice something's awry with R11 The board house decided to substitute this part for cost/availability reasons, didn't check the part # with us, and installed a wide 2010 resistor instead of a standard 2010. Then it passed visual inspection. Thank **** we didn't get them electrically tested, this resistor sets the maximum battery charging current and this would have blown the charger IC.
Right now we're trying to get the board house to own up to their mistake and fix it for us, provided they do that they'll be forgiven and remain nameless. I'd rework them except I'm on the wrong side of the atlantic ocean right now
On the up side, I've got one of the cards here with a reworked resistor and it's working great. Couple minor software tweaks to do (coming up with suitable "LED changes color at this battery voltage and DIP switch settings" type stuff), and once the card's reworked we've got ourselves a product for sale.
You may notice something's awry with R11
The board house decided to substitute this part for cost/availability reasons, didn't check the part # with us, and installed a wide 2010 resistor instead of a standard 2010. Then it passed visual inspection. Thank **** we didn't get them electrically tested, this resistor sets the maximum battery charging current and this would have blown the charger IC.Right now we're trying to get the board house to own up to their mistake and fix it for us, provided they do that they'll be forgiven and remain nameless. I'd rework them except I'm on the wrong side of the atlantic ocean right now
On the up side, I've got one of the cards here with a reworked resistor and it's working great. Couple minor software tweaks to do (coming up with suitable "LED changes color at this battery voltage and DIP switch settings" type stuff), and once the card's reworked we've got ourselves a product for sale.
Right now Viktor has a pile of assembled boards at his house
Standard version will support both HP-10W and GW-1058's. The current cAMP audio amplifier is a TPA3118 amplifier configured for 4 ohm loads, with a couple audio stages in front of it providing a highpass filter and differential receiver.
Provided you put a 4 ohm load (or thereabouts) on it, it'll work fine. 6/8 ohm loads will cause a bit of HF peaking.
can it do 2/3ohm?
I officially spec/guarantee the card for a 3.2 ohm minimum load. It should do 2 ohms, but you're pushing some things.
The output filter is set up for 4 ohm loads, a 2 ohm load will cause a bit of high frequency droop but nothing too serious.
The TPA3118 chip itself doesn't have a minimum load but rather an output current limitation - it's happy driving a 4 ohm load off a 24V supply, 2 ohms at 12V is the same current. Twice the output current means 4x the I2R loss in the TPA's output FETs though, it'll run hotter - it's got thermal protection so it shouldn't overheat and fail, but the audio will cut out if you reach this threshold.
Second limitation is the output inductor. I'm using Coilcraft JA4575-AL, which has this L-versus-I curve - Inductance vs Current - HA4158-EL, JA4575-BL Class-D Dual Inductors - it drops about 10% at 6 amps. You're pushing the inductor up into its saturation knee, which might cause a bit more THD/IMD, but the inductance droop doesn't look like it's enough to endanger the TPA3118. Thermally the part rises 20C at 4 amps, the crest factor of music listening shouldn't push the current that high.
So in theory you should be OK, I think if you're going to run into any problems they're going to be caused by the board overheating. To hopefully prevent that from happening, I'd suggest mounting the card with standoffs in an aluminum case, and using a thermal pad between the PCB and the case. I've actually provided an area on the bottom of the card for this guy:
H48-6-45-25-4-1A t-Global Technology | 1168-1701-ND | DigiKey
When you're done, test the heck out of it.
The output filter is set up for 4 ohm loads, a 2 ohm load will cause a bit of high frequency droop but nothing too serious.
The TPA3118 chip itself doesn't have a minimum load but rather an output current limitation - it's happy driving a 4 ohm load off a 24V supply, 2 ohms at 12V is the same current. Twice the output current means 4x the I2R loss in the TPA's output FETs though, it'll run hotter - it's got thermal protection so it shouldn't overheat and fail, but the audio will cut out if you reach this threshold.
Second limitation is the output inductor. I'm using Coilcraft JA4575-AL, which has this L-versus-I curve - Inductance vs Current - HA4158-EL, JA4575-BL Class-D Dual Inductors - it drops about 10% at 6 amps. You're pushing the inductor up into its saturation knee, which might cause a bit more THD/IMD, but the inductance droop doesn't look like it's enough to endanger the TPA3118. Thermally the part rises 20C at 4 amps, the crest factor of music listening shouldn't push the current that high.
So in theory you should be OK, I think if you're going to run into any problems they're going to be caused by the board overheating. To hopefully prevent that from happening, I'd suggest mounting the card with standoffs in an aluminum case, and using a thermal pad between the PCB and the case. I've actually provided an area on the bottom of the card for this guy:
H48-6-45-25-4-1A t-Global Technology | 1168-1701-ND | DigiKey
When you're done, test the heck out of it.
Last edited:
USBi clone
Great project
Just wanted to give a hint for anyone who still needs an USBi Interface
freeDSP | An Open-Source Low-Budget Audio DSP -> freeUSBi
regards
Ludwig
Great project
Just wanted to give a hint for anyone who still needs an USBi Interface
freeDSP | An Open-Source Low-Budget Audio DSP -> freeUSBi
regards
Ludwig
If you read our "freeUSBi getting started guide", you can see that AD actually provides everything.
We actually didn't build a USBi-clone, we just provide a simple THT-extension board to a cypress-board available on ebay from china.
And AD actually provides the USBi-Firmware with SigmaStudio, encapsulated within a script "... .spt".
But I don't think they would recommend to use anything besides the original USBi
In an update that we will put on our website quite soon, it's even easier. You only need do download a modified cypress-driver and the script provided with SigmaStudio, no additional software/tools required.
Sorry for this much half-offtopic within this nice project-thread
Almost ready!
Got sidelined for a little while by a Sony cellphone which appears to actively drive its audio ground to a different potential than the USB ground and made a hell of a racket when plugged into, and being charged by the card. This was going on over in Denmark, debugging things from here in Canada took a while
So with certain phones/MP3 players that pull this trick, the "ground loop breaker" circuit on the PCB might not be adequate and you may need a transformer isolator. Hopefully this isn't too common.
Got sidelined for a little while by a Sony cellphone which appears to actively drive its audio ground to a different potential than the USB ground and made a hell of a racket when plugged into, and being charged by the card. This was going on over in Denmark, debugging things from here in Canada took a while
So with certain phones/MP3 players that pull this trick, the "ground loop breaker" circuit on the PCB might not be adequate and you may need a transformer isolator. Hopefully this isn't too common.
Hey Everyone
Viktor from Canopy Sound here. Just wanted to let everyone know that our amplifier-batch has finally been tested and we are ready to ship them out during next week (we aim for wednesday)
Things got a bit delayed by exams and personal stuff (got a daughter two weeks ago) but we're finally where things looks ready now!
Right now we're researching the best over-seas shipping-mehtods for our US customers, so I hope we can have some decent prices for that as well.
Viktor from Canopy Sound here. Just wanted to let everyone know that our amplifier-batch has finally been tested and we are ready to ship them out during next week (we aim for wednesday)
Things got a bit delayed by exams and personal stuff (got a daughter two weeks ago) but we're finally where things looks ready now!
Right now we're researching the best over-seas shipping-mehtods for our US customers, so I hope we can have some decent prices for that as well.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.