SMPS design for power amps (Class D in particular)
I'm glad to see some action on the SMPS front. I've started this thread for continuing the discussion that was started in the UcD180 thread.
Are you an LTspice user? It works better than Pspice (IMO) and is a free download from the software page of the Linear Technology web site.
The simulation file I posted in the other thread uses averaged models for the switching stages, so it runs like lightning. The line current envelope is visibly modulated by the low frequency audio output, although this effect is lessened by the two pole low pass filter on the current feed forward.
If you haven't done so already, you ought to give LTspice a go with the pfc sim file from the other thread. It's fun. :joker:
Regards -- analog(spiceman)
Note: this post copied from the UcD180 thread for continuity
filter on the output current feedforward multiplier signal. Override operating limits are 4 amps on the input current and 400 volts on the output voltage. AC input range for proper operation is 80 to 264 volts
and nominal dc output is 375 volts. A 330uF capacitor is used for primary energy storage.
The pfc front end is followed by a simple dc to dc step-down supply for the isolated rail voltages. Its step-down ratio is 0.12, yielding a nominal 45 volts or so for the class d rails. A 10,000uF capacitor is
hung on each rail so that total output energy storage is just slightly less than that on the 330uF boost capacitor. The rail voltages are balanced by an auxiliary circuit that can easily be disconnected by
setting its series resistance to a high value.
The output load is my guess at an equivalent RLC network for a nominal 4 ohm woofer with a 34 Hz box resonance. For the test run shown, the amp is overdriven with a 15 Hz clipped sinewave burst. Without active balancing the positive rail voltage gets pumped up to a whopping 59 volts at the end of the burst.:umbrella: With the balancing circuit in place it remains safely well under 50 volts at all times.
An interesting note about the voltage balancing circuit: by designing it with a certain controlled level of impedance, it can avoid having to handle the large reactive currents required to keep the two output
capacitors in lock step, yet still nicely squelch the pumping effect. (I intend to cover my ideas on balancing circuits in another post.)
The LTspice simulation file is attached, but I had to change the file extension to from ".asc" to ".txt" in order to get it to post (you'll need to change it back after downloading).
Regards -- analog(spiceman)
Here is [the link to] the gif of the simulation: http://www.diyaudio.com/forums/attac...&postid=476737
I just figured out how link to an existing file which avoids wasting storage space on the server to duplications. My apologies to the group and moderators :cop: that I didn't think of doing this for the previously attached text file. :innocent: analog(spiceman)
Thanks for starting this thread, Bruno. Now I will take the opportunity to talk about my switching power supply design.
Anyone interested in circuit simplicity may like to investigate the USMPS. I thought of the name before I learned about the existence of the UcD. I like it as well as the UcD because it is composed of simple blocks. The USMPS is easy to modify for almost any application. I use variations of it for most of my power needs. Within the next several months, I plan to update the schematics and files on the Yahoo group I founded. If someone would like to become a group moderator, please contact me. That person should be passionate about the design and willing to promote it freely to all equally. I have no monetary profit motive.
Seems you're off to a good start!
Modelling a 4ohm woofer as having a 2ohm dc resistance is a bit unrealistic. The definition of "nominal impedance" is that at no point modulus of impedance is to drop more than 20% below this (3.2 ohms). The DC resistance is pretty much on that order.
15 Hz full power is somewhat unusual too, but it's a good acid test considering the loop bw of the PFC. I am under the impression (didn't test this - don't shoot me if I'm wrong) that the modulation is partly exacerbated by the PFC's own overshoot.
Could you add current feedforward to the simulation and check what happens then?
By the way, putting too much bandwidth in the current feedforward path most definitely does vastly increase the distortion and flicker in the ac input current waveform.
Regards -- analog(spiceman)
Good stuff analogspiceman.... :)
Alternative SMPS Configurations
Hi Guys, and thanks Bruno for starting this thread. I have also been following the UcD thread with great interest.
I think an SMPS for amps is of interest to a lot of people and it is a subject I have spent some time researching myself. The traditional approach to include PFC seems to be the standard PFC boost converter front end followed by some form of switching down converter (half bridge/full bridge, etc).
One possible alternative that I have looked at but is rarely seen commercially is the current fed topology (inductor in the primary line). This may be a push-pull, full bridge, etc and allows the PFC and down conversion to be achieved in a single stage. Because a PFC IC is used in the control loop secondary ripple rejection is low and will be determined by the size of capacitor tank on the secondaries but a large bank is desirable for audio amplifiers particularly in class-d half bridge because of supply pumping. However, although there are large caps in the secondary there aren't in the primary (because of the pfc action).
What this circuit does is achieve output power regulation (albeit with some mains line ripple) over a universal input voltage range and is scalable from small powers to kilowatts where it is most cost effective.
To summarise, the following advantages and disadvantages of this topology are:
PFC and down conversion in a single stage.
Magnetic components approximately the same as the two stage approach. (But still considerably reduced in size over a normal 50Hz supply)
Excellent cross regulation characteristics on multi output supplies because of the single primary inductor current feed with no minimum output current requirements like a normal forward mode PSU.
Secondary regulation across a wide supply input.
No shoot through when configured as a bridge or push pull because of the current fed feature - in fact part of the operation as you store the energy in the inductor is through deliberate overlap.
Increased output mains ripple over the 2 stage approach but for an amplifier with normal PSRR this is not a problem.
Slightly increased high frequency output noise but can be oversome with some secondary fillering.
No off the shelf control IC's - but can easily be implemented with a conventional PFC ic with some external logic.
Possibly have to use an active primary clamp depending on design configuration.
After doing quite a bit of work on this it looks very attractive for cost effective small size audio supplies with integral PFC, however I have never seen it used commercially. This maybe because of some inherant problems (which I would be interested in comments on) or alternatively because not much has been written about this configuration.
I look forward to some feedback
Re: Alternative SMPS Configurations
I've finally found a small amount of funding for looking into a single-stage pfc smps, albeit using a somewhat different topology. I have once seen a paper that does what you suggest (if I understand you well). A pair of switches build a pp converter. The switches are made to overlap to charge a coil which is in series with the ct of the primary. There could be a controller IC to do this.
The thing I'm looking at is: replace the diode in the pfc boost converter by a small switch and change the boost cap into a small film thing (now an active clamp). Make the boost coil the primary of a transformer. Rectify the secondary with an inductor-input fw rectifier.
The voltage on the active clamp cap is a measure of the secondary voltage so that can be used as a feedback takeoff point.
Performance is identical to a two stage converter.
I can start this thing end of november. I'll keep you posted.
|All times are GMT. The time now is 12:48 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright ©1999-2015 diyAudio