• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

BOSC - HiFi Monoblock Class-D GaN Amp

This is absolutely bonkers (in a positive way).

Thank you.

Yes, the Purifi has better measurements, but it is also a significantly more complex topology (three nested feedback loops, and one feed forward loop crossing a feedback point).

If you compare THD at 1W BOSC clearly is the winner especially at high frequencies. Compare figure 7 to figure 10.
 

Attachments

  • Figure7.JPG
    Figure7.JPG
    57.8 KB · Views: 412
  • Figure10.JPG
    Figure10.JPG
    49.7 KB · Views: 408
No. You have to provide it externally.

There's absolutely no way anyone can provide cycle by cycle current limiting on the MOSFETs externally.

What do you want them to do. Hack up the PCB, install current sensing resistors, and then tune additional analogue circuitry in an attempt to protect the MOSFETs? How are you supposed to test it works? Potentially blow up your amp by accident?

How would you suggest adequate protection against accidental short circuit and over current should be added?
 
There's absolutely no way anyone can provide cycle by cycle current limiting on the MOSFETs externally.

What do you want them to do. Hack up the PCB, install current sensing resistors, and then tune additional analogue circuitry in an attempt to protect the MOSFETs? How are you supposed to test it works? Potentially blow up your amp by accident?

How would you suggest adequate protection against accidental short circuit and over current should be added?

The power supply that I provide with the amps provides overcurrent protection, I have extensively tested and it is fast enough to protect the MOSFETs. It does not have to be cycle by cycle, just fast enough.
 
That's not a solution for those buying the amplifier boards or building multi channel amps built around a single supply.

For $375 the amplifier itself should have protection included.

If you build a GaN amp around TIs other FETs with integrated drivers some of those come with built in protection. It's a massive oversight, in my opinion, that TI didn't include current limiting in the half bridge.
 
I am sure that such protection will not work if you close out the amplifier's output at full power. Test this on video for those who doubt

Really, this amp is very good, you use nCore structure? SOA+integrator?

It is my own proprietary dual feedback modulator. There is additional info in the AudioXpress article I posted earlier, take a look.
 
Ex-Moderator
Joined 2005
Nice project, looks like the integrated TI GaN half bridges?

Are you're using the TI half bridges? I had posted about using the TI LMG5200 over 5 years ago (still don't have the time or resources to attempt what you did) but no one else seemed excited about it - TI LMG5200 half bridge power stage

Glad to see someone finally integrated one of those power stages into a hifi amplifier! The design looks nice and compact and thoughtfully partitioned (from what the photos show). Well done!

You had mentioned you chose a more expensive, non-standard stack up to reduce inductance. Why didn't you just use a more standard 8 layer stackup to reduce cost while still reducing inductance? And did the use of a "higher quality" laminate really provide any audible difference? Sorry for all the questions, I missed all this fun while on hiatus.
 
Are you're using the TI half bridges? I had posted about using the TI LMG5200 over 5 years ago (still don't have the time or resources to attempt what you did) but no one else seemed excited about it - TI LMG5200 half bridge power stage

Glad to see someone finally integrated one of those power stages into a hifi amplifier! The design looks nice and compact and thoughtfully partitioned (from what the photos show). Well done!

You had mentioned you chose a more expensive, non-standard stack up to reduce inductance. Why didn't you just use a more standard 8 layer stackup to reduce cost while still reducing inductance? And did the use of a "higher quality" laminate really provide any audible difference? Sorry for all the questions, I missed all this fun while on hiatus.

Yes, they are TI half-bridges.

The stack-up and dielectic work together to minimize any parasitic inductances and capacitances especially when it comes to the switch node, because the slew rate of GaN devices is incredibly fast.