-
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!
Class D Amplifiers 101
- Thread starter DonH56
- Start date
All this to support 4 itsy bitsy GaN chips - or maybe it's 8 (big difference?).
Tks
Major Contributor
- Joined
- Apr 1, 2019
- Messages
- 3,221
- Likes
- 5,496
One thing I'll give them from the photo's.
Sure is one of the most beautiful pieces I've seen.. Not gaudy, or as is usual in the high-end where you have guady, but sometimes you notice the finishing is pretty shit. (You have to be real anal to see the differences, like watch-movement finishing sort of anal).
Edit: Moved related, but off-topic amp modification post to separate, new thread.
Last edited:
The Psemi vid mentions parasitic diodes in standard MOSFETS - those are actually body diodes:
https://en.wikipedia.org/wiki/Power_MOSFET#Body_diode
Actually useful as a freewheel diode in some applications...
https://en.wikipedia.org/wiki/Power_MOSFET#Body_diode
Actually useful as a freewheel diode in some applications...
OP
- Thread Starter
- #30
I did not look at the link, not sure it's relevance to this thread, but some power transistors (FETs, BJTs, etc.) include explicit clamp diodes. They may be, but often are not, body diodes but rather extra diodes, either fabricated on the device die, or discretes added inside the package.
From that link:
Body diode
It can be seen in figure 1 that the source metallization connects both the N+ and P+ implantations, although the operating principle of the MOSFET only requires the source to be connected to the N+ zone. However, if it were, this would result in a floating P zone between the N-doped source and drain, which is equivalent to a NPN transistor with a non-connected base. Under certain conditions (under high drain current, when the on-state drain to source voltage is in the order of some volts), this parasitic NPN transistor would be triggered, making the MOSFET uncontrollable. The connection of the P implantation to the source metallization shorts the base of the parasitic transistor to its emitter (the source of the MOSFET) and thus prevents spurious latching.
This solution, however, creates a diode between the drain (cathode) and the source (anode) of the MOSFET, making it able to block current in only one direction.
Body diodes may be utilized as freewheeling diodes for inductive loads in configurations such as H-bridge or half bridge. While these diodes usually have rather high forward voltage drop, they can handle large currents and are sufficient in many applications, reducing part count, and thus, device cost and board space.
In fact most of the schematic symbols I use in my designs show that...
As to relevance, the
https://www.psemi.com/newsroom/pres...fastest-switching-speeds-to-gan-class-d-audio
the page mentions:
“GaN is disrupting traditional, power MOSFET markets and changing the way we live,” says Alex Lidow, CEO and co-founder of Efficient Power Conversion Corporation (EPC). “In class-D audio systems, the audio performance is impacted by the FET characteristics. Our enhancement-mode GaN (eGaN®) transistors enable a significant increase in the sonic quality and higher efficiency. High-speed FET drivers, like Peregrine’s PE29102, are critical to unlocking the performance potential of eGaN FET technology in applications like class-D audio.”
In the video that's on that page:
Mentions the "parasitic diode" - also known as "body diodes- as well as gate capacitance as a source of jitter in typical MOSFETS:
OP
- Thread Starter
- #32
Yes, yes, you can assume a basic level of device physics on my part, I am just not sure how this ties to what is essentially a very high-level introduction to class D operation? I intentionally did not use real devices to make it easier to follow, easier for others to recreate the simulations, and to avoid discussing device physics and low-level circuit theory. For the most part my assumption is that folk who actually do the designs know it all already and would not learn anything from these various introductory threads (but hopefully will read anyway and correct any mistakes I have made), and folk who are not designers probably would not be able to follow a discussion at the transistor level (or below). And the parasitic diodes, body and source/drain, are one step below just understanding the transistor itself, so it seems you are going deeper than required for a thread of this sort.
Clamp diodes are in a lot of power devices, at least last I looked at them (not my day job), but were often separate diodes in a hybrid package so they would clamp at a lower voltage before body or S/D diodes biased fully on (like Schottky clamping diodes).
I must be missing your point, sorry, long week... - Don
Clamp diodes are in a lot of power devices, at least last I looked at them (not my day job), but were often separate diodes in a hybrid package so they would clamp at a lower voltage before body or S/D diodes biased fully on (like Schottky clamping diodes).
I must be missing your point, sorry, long week... - Don
Falstad has a great simulation of the basic Class D amp:
https://www.falstad.com/circuit/e-classd.html
also here:
http://lushprojects.com/circuitjs/
Just go under the Circuits menu > Opamps > Amplifiers > Class D
You can add components, view nodes of the circuit in a scope (just right-click on the wire). For instance, right click on the left of the 80mH inductor and add a scope - you'll see the signal before the inductor filters it
To add wires and components, just right click on a blank area...
For fun you can muck around with the triangle freq - just right click the 1kHz triangle generator. You can change it to other waveforms too.
Download for offline:
https://www.falstad.com/circuit-java/circuit.zip
Un zip and double click on the circuit.jar file
https://www.falstad.com/circuit/e-classd.html
also here:
http://lushprojects.com/circuitjs/
Just go under the Circuits menu > Opamps > Amplifiers > Class D
You can add components, view nodes of the circuit in a scope (just right-click on the wire). For instance, right click on the left of the 80mH inductor and add a scope - you'll see the signal before the inductor filters it
To add wires and components, just right click on a blank area...
For fun you can muck around with the triangle freq - just right click the 1kHz triangle generator. You can change it to other waveforms too.
Download for offline:
https://www.falstad.com/circuit-java/circuit.zip
Un zip and double click on the circuit.jar file
Last edited:
While GaN FETs can be operated up to 100A, 200V, and 100MHz switching frequency, still, no class-d amp either Silicon-based or GaN-based, ever exceeded the 1MHz switching frequency barrier due to diver limitations and output low pass filter for high design problems for very high switching frequency (no inductor can handle more than 1MHz, due to stray capacitance and non-linearity and core losses). GaN has many advantages. No body diode, meaning no reverse recovery issues, lower Rdson resistance, meaning infinite damping factor and near 99% efficiency, and low outout/input capacitance (Coss), meaning very high switching frequency, which means virtually zero harmonic distortion.
Once ultra-high frequency inductors and gate drivers become available, all audio amps will be based on GaN class-d. A small lightweight class-d amp can deliver 2000W continuous power into 2 ohms without the need of a heatsink if based on GaN FETs.
It will take another decade for a 2000W 0.00001% THD 120db SINAD class-d amp, which easily outperforms D'agostino's relentless 250,000$ 400lb crap amp, to become available on Aliexpress for about 200$...
Once ultra-high frequency inductors and gate drivers become available, all audio amps will be based on GaN class-d. A small lightweight class-d amp can deliver 2000W continuous power into 2 ohms without the need of a heatsink if based on GaN FETs.
It will take another decade for a 2000W 0.00001% THD 120db SINAD class-d amp, which easily outperforms D'agostino's relentless 250,000$ 400lb crap amp, to become available on Aliexpress for about 200$...
OP
- Thread Starter
- #35
Air-core inductors are used in multi-GHz RF amplifiers and high-power oscillators... There have been T-coil-like approaches (or duals of them) proposed to help compensate inductor parasitics. Also interleaving schemes that ideally suppress HF noise by running output stages in quadrature and such. Not something I have followed closely, however.
2 kW without a heat sink would be sporty even for class D, seems like you'd need 99% efficiency or better to get the power below that needed to operate with no heat sink at all (1% of 2 kW = 20 W). I haven't seen anything anywhere close to that yet but again not something I track closely.
2 kW without a heat sink would be sporty even for class D, seems like you'd need 99% efficiency or better to get the power below that needed to operate with no heat sink at all (1% of 2 kW = 20 W). I haven't seen anything anywhere close to that yet but again not something I track closely.
ZeDestructor
Active Member
- Joined
- Jul 28, 2019
- Messages
- 119
- Likes
- 68
Given the size constrainsts of the FETs and their packaging, you can't go with more than ~3W of waste heat if you want to run without a heatsink at all. Anything higher, you need airflow, heatsinking, or airflow through a heatsink. A few GaN-based PC PSUs have showed up in recent times doing over 93% overall efficiency at 1600W, and even those run heatsinks on the primary-side GaN MODFETs despite having a large 140mm fan to cool the whole thing (and the primary side achieving well over 97% efficiency), just because of how cramped the space is, despite being kept very far from the secondary-side MOSFETs
MOSFETs and IGBTS efficiencies are getting better all the time... as they lower the on resistance, they generate very little waste heat...
I use an Optomos that's got 50mohm RDSon (on resistance) and is rated at 9A DC with no heat sink - 4 lead 264 case style. We run these at about 30VDC so that's about 270 watts at the full 9A. One application runs about 6A and with continuous DC we get very little (only see it on a FLIR) temp rise - that's 150 watts...
https://www.digikey.com/product-det...-circuits-division/CPC1709J/CLA280-ND/1277135
http://www.ixysic.com/home/pdfs.nsf/www/CPC1709.pdf/$file/CPC1709.pdf
There's an AC version of their Optomos in a low profile package with no heatsink - 6A RMS 60mohm RDSon
https://www.digikey.com/products/en?keywords= CPC1907B
http://www.ixysic.com/home/pdfs.nsf/www/CPC1907B.pdf/$file/CPC1907B.pdf
I talked to the one FAE (field app engineer) at CP Clare (INXYS) and he mentioned they're working on some stuff that's even better... I think they and Maxim have aliens over there...
Technics SE-R1 amp switches at 1.5MHz and does direct digital to pwm conversion: https://www.technics.com/us/products/r1/se-r1.html
I believe EPC has a demo board which can be used as a class D amp with a driver capable of 2MHz operation.
I'm not sure what switching frequency Orchard Audio is using: https://orchardaudio.com/bosc
I’m not in my lab at the moment to check my data for sure, but I seem to recall that it was around 750kHz.
Looks like 800kHz as per the last foil of this test report
https://drive.google.com/file/d/1IZeZp85TPiGTmx6GkCuusMb_tQBoL--C/view
See the last page FFT with source signal of 1kHz and carrier signal at 800kHz (could be 750kHz)
Similar threads
