Class D modulator shoot-out

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2004
Paid Member
This experiment is in the early stages, but I've got some of it working and decided to post a teaser. This amp allows switching between two different modulators for the same output stage (STA317B). I've got the analog signal path "sort of" working. Lots more to do.

Now I wish I had included a TI modulator on this board as well.

apogee.jpg
tripath.jpg
 
Member
Joined 2004
Paid Member
Interesting....

Are you using the Tripath TC2001?

TC2000 and DDX8001 (STA308A). Uses the Cirrus CS42526 to convert the analog to digital for the Apogee chip and to convert SPDIF to analog for the Tripath chip. Also has 3 30A relays per channel to switch in different output filter components to compensate for the different switching frequencies.

No real comparisons yet, and I'm just using a small power supply for testing, but the big obvious difference is how much quieter the Apogee chip is.
 
Member
Joined 2004
Paid Member
What was the motivation to build a complicated setup to switch modulators and outputs filters as opposed to building two separate amplifiers?

The amp has an IR remote control sensor, so it will be easy to A/B. This approach allows using the same speakers, which is nice. The relays also allow selecting between 4ohms and 8ohms, using the recommended values from ST/Apogee.

There are other reasons, as well...

Is STA317B the correct part number for the output stage; or did you mean STA517B? Google doesn't return any results for STA317B.

Sorry--should be STA517B. 48V supply
 
Member
Joined 2004
Paid Member
What is your overall impression of the TC2000 + STA517B?

I have ~100pcs of TC2001 just begging to be put to use.

I'm still getting ready to do some serious evaluation. I've got the A/D working for the Apogee and the D/A for the Tripath, and I can send digital directly to the Apogee modulator (SPDIF) and analog directly to the Tripath. I can switch back and forth between the two modulators and the output filter switching is working. I can control all of the DSP and volume controls in the Apogee...so a lot of the control stuff is done. However, I need a good quality digital volume control for the Tripath modulator--still working on that.

One of the "gotchas" for the Tripath is that the STA517B requires a change in the feeback resistors to accomodate the higher voltage. However, this change increases the modulator gain of the amp by about 4dB. The Tripath input op amp is not super quiet, so there is a lot of hiss: I need to change the gain on the input op amp.

Just from initial listening with a budget source and crappy speaker, the Tripath and Apogee modulators appear comparable. Right now the obvious difference is the background noise: lots of hiss with the Tripath modulator and dead silence from the Apogee chip (zero detect is enabled).

Real testing to start in about two weeks. Will be testing a number of power supply options as well (350 switcher vs unregulated supply, +capacitor bank, +LDO regulator).
 
Administrator
Joined 2004
Paid Member
Great stuff Neil! Looking forward to hearing more about the project.

I agree about the Apogee chips. Not a hiss, buzz, or anything. Amazingly quiet.

Some time last year I made a wave file with a tiny signal so that the Apogee chip would not shut down. It was still just as quiet.
 
Member
Joined 2004
Paid Member
Great stuff Neil! Looking forward to hearing more about the project.

I did some fairly extensive testing today with a wide range of music at various volume levels. Got my son and wife involved, as well.

I just used the digital path from the CD player, going through a good-quality Cirrus DAC to the Tripath modulator and with the digital signal routed directly to the Apogee chip (see the blue path in the first picture).

My son preferred the Tripath, but he only listened at low volumes and probably has "suspect" listening tastes due to growing up on MP3's. My wife and I listened during dinner and we both agreed that on percussive music the Apogee modulator sounded much more "open" and "clear". The Tripath has a warmth that sounds better on piano but doesn't work well for drums and other music with sharp attacks. They are both good amps, and it's hard to tell the difference on a lot of music. Of course, at really low levels the Apogee is silent, whereas the Tripath has some background hiss. We agreed that the Apogee was probably the better modulator overall.

Another point: the STA517B H-bridge running at 48V provides enough power to be "fun"--it's the right amount of power to fill a large room

signal_flow.gif

The amp is pretty cool--very small and it runs fairly cool, with heatsinking to the chassis. It's got output filter switching for 4/8 ohm speakers, and everything is switched using an IR remote control. When the modulators get switched it automatically adjusts the volume so the modulators are at the same volume level.
Tripathogee.jpg

The amp also has connectors to evaluate various types of power supplies. The power supply board isn't done, but it will allow switching between a switching power supply and a linear supply, and it has relay switches to use a 60,000uF filter bank or an LDO regulator. The chassis area on the upper right is for the transformer and power supply board
Tripathogee_chassis.jpg

This is one of those designs you do just to learn things. Now I have to fix a few more things and then make another one so other people can evaluate it.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.