"...contribute to a highly dynamic sound image with ultimate time correctness...."
Marketing blabla.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
GaN Systems Amplifier Eval Board Measurements
- Thread starter amirm
- Start date
Marketing blabla.
Prevent vibration caused failure.
Speaking specifically of cell phones: in the quest for efficiency some seem to have forgotten the quality of the sound that the microphone and the speakers should ALSO BE GREAT (clear & loud enough, as should the ringers, with the background noise in my work [which is somewhat loud but does not require earplugs] unless the phone is in my top pocket (a very precarious place for a phone to be), I won't hear it. And with the work that I am doing, I may or may not feel it vibrate. Holding even a short conversation on it is tedious, at best. Ear buds are not aloud due to that you may not hear the danger of a forklift or some other hazard. In general, we haven't gotten there yet...
The MOSFETS in my amp are $10 apiece, with a full bridge design you are looking at $40 per amp.
A regular MOSFET is around $2.
So right there you are looking at a difference of $32 per amp.
eliash
Senior Member
...btw., just repaired the switched motor control unit in our ~30y old Siemens washing/dryer machine, a large foil cap unsoldered itself due to ultrasonic vibration and associated micro-arcing...glue could have helped...
Last edited:
eliash
Senior Member
These GaN devices could also be used to push up the switching frequency of classD amps by some factor 2, thus allowing a better (power) bandwidth...and come closer to good classAB designs...next gen Purify???
Higher output switching frequency also adds strain to the gate driver. The dissipation goes up exponentially as frequency goes up. Another issue is class D should avoid AM frequency range. This means you either stuck at 400+khz range or you need to push over 1.7MHz. I wouldn't say it's impossible but each time the switching frequency being doubles it's a new level. Not only heat, dead time management and its consistency and production variation becomes critical. Or it's better to reduce the frequency to avoid shoot through or distortion increase.
What's needed to get to the next chapter actually isn't distortion but noise. The complicity of the designs often adds noise. Loop gain reduces distortion but noise is only dependant to closed loop gain and input noise.
Staying below MW is however not a rule of thumb
It seems we can see usual excuses that will be left after the class D technology is getting improved. There is a lot of space for improvements, still.
And AM band is affected even in case of 400kHz carrier as there are always harmonics. So let's not fool the members.
OP
eliash
Senior Member
Long and medium wave transmission has been given up here in Germany for some years now, other European countries will follow...and less gate capacitance influence is relaxing the driver requirements...but yes beyond 1.7MHz is the safe side for switching, spectrum already occupied by PV converters everywhere here...next is EV charging...no fun for radio amateurs...
Last edited:
restorer-john
Grand Contributor
Let's face it, LW, AM and even SW is from an era gone by. I loved DX-ing back when I was young. Incredible times picking up capital city (Australian) stations on AM 1500-3000km away from me at night, or going for regional stations later at night when everything was quiet to get a country town out in the middle of our huge country. Going LW and SW to listen to VOA (voice of America) out of Guam or dialling in UK, Russian, Chinese or US stations. It was wonderful and I will never forget it.
Now AM is a joke. ERP is down from 250kW+ Tx to just enough to cover a small area. FM rules on radio and even that is rubbish.
Sad, but progress I guess.
The great sadness about the demise of AM is that soon it will be useless to construct a crystal set receiver. There is little to beat the wonder of assembling a few simple components and sound comes out. You can make almost all the components from scratch which adds to the wonder.
Sometimes I feel like running a petition to preserve a station for just this reason.
Sometimes I feel like running a petition to preserve a station for just this reason.
restorer-john
Grand Contributor
A slug/ferrite tuned coil, a ceramic cap and a germanium diode, along with a salt-crystal earpiece and any bit of wire you could clip an alligator clip to. Perfect.
Luckily our local AM station was only a few km away and it was pumping out some serious power. My Dad's phono stage and speaker leads made for a nice AM reciever- much to his disgust and annoyance.
Or if you were a fancy hobbyist with a ball bearing, brass shafted, variable cap in a beautiful metal frame, screwed down to whatever piece of wood (mum's old chopping board) you could scavenge. Nothing in the world was more joyous to me than a radio station, in my ears, with no active circuitry at all. Magical. No batteries required.
Last edited:
eliash
Senior Member
(now we have the ideal Sunday theme here)..maybe that´s why some of us are kind of hesitant about classD amps and such SPS', knowing about what it means to listen to Newfoundland St. John's AM station in western Europe at night or the news via federal medium power AM station from Germany during daytime on some Maltese island...now there is only Vinyl left, where the faint sound of some 10nm stylus excursion matters...
It makes me smile to be able to read some coherent text on a possible new improvement in semiconductor applications. As to increased switching speed, it seems to me that the further we can get between the highest audible frequency and the power amp switching frequency, the better. It's about much more than just efficiency. Output filtering just has to be easier, inductors and caps smaller. That said, we are probably fairly early in the practical design cycle, so I don't expect to see them in my equipment cabinet soon.
<chuckle> We probably need to prevent these switching frequencies from transmitting to submarines at sea.
-Just one man's view
<chuckle> We probably need to prevent these switching frequencies from transmitting to submarines at sea.
-Just one man's view
Trinitrond
Member
- Joined
- Feb 21, 2021
- Messages
- 8
- Likes
- 3
Technics/Panasonic use their own developed Gan Transiators in SE-R1, SU-G30 and the new SU-R1000.
Technics use their own jitter noise reduction tech with battery powered clock to reduce noise.
I guess that is one of the keys to improve snr.
A little heatsink can also improve GaN performance as stated in several White papers on GaN.
Heatsinks are not free either so you have some cost saving there.
Product margin is maybe the "highest" cost in audio Vs It-industry as volume is lower to recap development cost.
Technics use their own jitter noise reduction tech with battery powered clock to reduce noise.
I guess that is one of the keys to improve snr.
A little heatsink can also improve GaN performance as stated in several White papers on GaN.
Heatsinks are not free either so you have some cost saving there.
Product margin is maybe the "highest" cost in audio Vs It-industry as volume is lower to recap development cost.
True in every industry. In IT people are just used to the commodity parts. Billions of dollars to develop a next gen processor, billions of dollars to develop and build the next gen fab. Huge volumes and ridiculous performance for the money. But at the pointy end, there is plenty of very low volume very high end compute with jawdropping prices (and capabilities.) A single top end FPGA chip is in the tens of thousands. By the time it gets into a product the prices are eyewatering. And there is everything in between.
eliash
Senior Member
...actually, I remember, when I was still in school, I tried almost exactly this suggestion:
"A slug/ferrite tuned coil, a ceramic cap and a germanium diode, along with a salt-crystal earpiece and any bit of wire you could clip an alligator clip to. Perfect. "
It did not work!
Why?: The crystal-speaker is high impedance, causing a rectified voltage build-up at the high-impedance speaker, resulting in shifted diode operational point and no sound. I did not understand this fully at that time (it would have been so easy to add a parallel resitor to the crystal-speaker), but fortunately the conducting dynamic headphone did the job...
The total power needed to drive a MOSFET gate is:
Driving a MOSFET a typical gate drive voltage will be 10 to 12V, while driving a GaN device the gate driver voltage will be 5 to 6V.
Another thing to factor in is that the total gate charge for GaN transistors will be lower than that of an equivalent MOSFET.
In summary, even though the switching frequency goes up by a factor of two the power needed to drive the gate will still be lower than for a MOSFET at a lower frequency.
Here is a very good app note on the subject:
https://www.ti.com/lit/an/slyt664/slyt664.pdf
For my design the total gate drive power needed is about 65mW per GaN FET x 4 = 260mW. I have verified this by measurement.
Last edited:
Similar threads
- Poll
