LLC resonant SMPS's are too expensive for powering Class D guitar amplifiers...because that market is too small, and is too low-cost driven, for LLC resonant converters to be suitable
do you agree?
LLC smps's are only for big production runs, or exotic applications like radar PSU'S?
-do you agree that the resonant capacitor has to be a well toleranced high voltage component, and such components are too expensive for small production runs......and a plain half-bridge would be a better option?
Also, the resonant inductor ("leakage") has to be very tightly toleranced, and whether its a separate inductor, or part of a "divided bobbin" transformer, its very expensive....too expensive for small volumes for cheap power supplies.
As stated, surely the resonant frequency has to be pretty well defined?
...so i would conclude that LLC resonant converters are either for exotic applications in small runs (eg radar PSU's) -or for high volume production runs where the tightly toleranced components become cheaper due to the high volumes....is this right?
do you agree?
LLC smps's are only for big production runs, or exotic applications like radar PSU'S?
-do you agree that the resonant capacitor has to be a well toleranced high voltage component, and such components are too expensive for small production runs......and a plain half-bridge would be a better option?
Also, the resonant inductor ("leakage") has to be very tightly toleranced, and whether its a separate inductor, or part of a "divided bobbin" transformer, its very expensive....too expensive for small volumes for cheap power supplies.
As stated, surely the resonant frequency has to be pretty well defined?
...so i would conclude that LLC resonant converters are either for exotic applications in small runs (eg radar PSU's) -or for high volume production runs where the tightly toleranced components become cheaper due to the high volumes....is this right?
don't think so. LLC is becoming a mainstream topology for 100W...600W, with good results. as the current is almost sinusoidal, the EMI is pretty good and noise is small. And, with the resonant inductor being inside the feedback loop its precision does not really matter so much as long as the converter stays inside appropriate frequency ranges for the whole output power range. I prefer a separate inductor for its much smaller magnetic leakage field.
just my two cents...
just my two cents...
Thankyou hesener, i agree with everything you say, whihc is excellent adive, but the fact still remains, that LLC is not for small production runs of cheap power supplies................for the reasons i have stated.
With LLC , you need to be operating pretty much on the upper resonance frequency, and with a resonant cap of tolerance +-20%, same for the inductor etc, , you cannot be sure you are near enough to the resonant frequency.
With LLC , you need to be operating pretty much on the upper resonance frequency, and with a resonant cap of tolerance +-20%, same for the inductor etc, , you cannot be sure you are near enough to the resonant frequency.
Hello,
Hesener, do you have a good experience on LLC?
I have built a prototype during the last years, but I am not too happy with the results.
First, I have to review my calculations because the mosfets are hard driven. About 4-5 times, they exploded in my face. I guess that the converter was working in the capacitive region.
The regulation loop need much more attention. I have found equations in the AN2644 on the L6599 from ST.
Otherwise, I like the principle.
Thanks for your feedback.
Cheers,
Serge
Hesener, do you have a good experience on LLC?
I have built a prototype during the last years, but I am not too happy with the results.
First, I have to review my calculations because the mosfets are hard driven. About 4-5 times, they exploded in my face. I guess that the converter was working in the capacitive region.
The regulation loop need much more attention. I have found equations in the AN2644 on the L6599 from ST.
Otherwise, I like the principle.
Thanks for your feedback.
Cheers,
Serge
Serge i hear you say the fets exploded..this confirms exactly my point so i thank you.
Your resonant parts were probably wide tolerance...and bang!
This is the problem with the LLC topology.
It is simply not for cheap PSU's in small volumes.
Also, where it is done.........terrible thing to say, but it's often a power supply for the engineer's Curriculum Vitae.
Your resonant parts were probably wide tolerance...and bang!
This is the problem with the LLC topology.
It is simply not for cheap PSU's in small volumes.
Also, where it is done.........terrible thing to say, but it's often a power supply for the engineer's Curriculum Vitae.
Last edited:
control loop design on the LLC is not easy, I agree with that. It is crucial to have a switching frequency high enough to be above the lower resonance, so that the gain decreases with increasing switching frequency.
check AN-4151 from Fairchild, it should give you a solid (mathematical) understanding of what is involved here. check out the diagram on page 3 lower righthand corner. was the primary inductance of your transformer appropriate, or too high maybe? difficult to move forward without a concrete design.....
just my two cents
check AN-4151 from Fairchild, it should give you a solid (mathematical) understanding of what is involved here. check out the diagram on page 3 lower righthand corner. was the primary inductance of your transformer appropriate, or too high maybe? difficult to move forward without a concrete design.....
just my two cents
Thanks Guys,
I'll check the Fairchild AN-4151.
I do not remember the inductance. I computed the turn according to the formula and the typical leakage (uH/squared turns) of the transfo. The project has been shelved for the last two years. I found an excel file that does the computing. I'll have to find it and post it or the link here.
Thanks again.
Serge
I'll check the Fairchild AN-4151.
I do not remember the inductance. I computed the turn according to the formula and the typical leakage (uH/squared turns) of the transfo. The project has been shelved for the last two years. I found an excel file that does the computing. I'll have to find it and post it or the link here.
Thanks again.
Serge
There's another consideration, and excuse me if I am preaching to the choir.
Many of the best guitar amps are not accurate reproducers, but have very special sound-generation characteristics when operated outside the "normal linear" range. Power supply sag is one component of the compression and distortion which makes the electric guitar so emotive; the amplifier's voice and timbre may change dramatically with the guitar's incredible dynamic range; the player can go from clean and smooth croon to hard biting scream with just a harder pick attack or bump of the guitar's volume knob. The guitar itself has incredible dynamic range, but the amp can either compress to make every note the same, a form of auto-correction for sloppy hammer-on, pull-off, sweep-picking technique etc. or can use the guitar's dynamic range as a "controller" for emotive tone changes, like a saxophone or human voice does when driven hard.
Now, for orchestral F-hole guitar, some forms of Jazz, some bass guitar, and pedal steel the requirements may be for accurate reproduction instead of such an active part of the sound creation process. Or the processing of dynamics, tone, and effects can be handled primarily by the preamp itself or digital emulation before the power amp; sometimes that pre-processing results in a signal to the power amps which is conveniently pre-compressed, which is near-ideal for a SMPS and class D. Remember that a very loud "clean" punchy sound preserving dynamics and attack is also hard to do with a class "D" and/or SMPS unless you "oversize" the amp considerably. A normal power supply and amp can be designed to handle short-duration (a few ms) demands via very large oversized power supply caps, and reproduce a short dynamic peak that is 100 times its continuous rating or much more (the opposite of power supply sag). A SMPS with normal small caps sized as high-frequency ripple filter and/or a class "D" amp may have wonderful continuous rating and sound great up to that point, but no additional dynamic headroom at all, and require input compression to avoid some really bad sounds. For playing punchy bass or clean guitar, when I use a class "D" or SMPS amp I require several times more continuous power rating in order to always operate within its normal dynamic range, in which case it works very well for a clean sound. For instance, for a clean punchy bass gig I might take two Peavey classic 120 power amps with 4 6L6 in each, for a total of 120 watts continuous clean (though Peavey claims that for each), or two Peavey Deca 1200 SMPS Class D power amps, for a total of 2400 watts (no need to say "continuous" because the continuous and peak ratings are the same). Size and weght is similar. The tube amps can haandle short peak demands or be pushed beyond their limits and sound very musical. The SMPS/"D" amps must always remain within their operating limits and have their limiter switches turned on.
On the other hand, they've started making some very compact very clean-sounding "boutique" pedal steel amps using very high-powered SMPS class "D" amps built into a combo-size speaker box.
All I'm trying to say it that making a guitar amp you or someone else will love must have equal parts art and science for mutual support.
Many of the best guitar amps are not accurate reproducers, but have very special sound-generation characteristics when operated outside the "normal linear" range. Power supply sag is one component of the compression and distortion which makes the electric guitar so emotive; the amplifier's voice and timbre may change dramatically with the guitar's incredible dynamic range; the player can go from clean and smooth croon to hard biting scream with just a harder pick attack or bump of the guitar's volume knob. The guitar itself has incredible dynamic range, but the amp can either compress to make every note the same, a form of auto-correction for sloppy hammer-on, pull-off, sweep-picking technique etc. or can use the guitar's dynamic range as a "controller" for emotive tone changes, like a saxophone or human voice does when driven hard.
Now, for orchestral F-hole guitar, some forms of Jazz, some bass guitar, and pedal steel the requirements may be for accurate reproduction instead of such an active part of the sound creation process. Or the processing of dynamics, tone, and effects can be handled primarily by the preamp itself or digital emulation before the power amp; sometimes that pre-processing results in a signal to the power amps which is conveniently pre-compressed, which is near-ideal for a SMPS and class D. Remember that a very loud "clean" punchy sound preserving dynamics and attack is also hard to do with a class "D" and/or SMPS unless you "oversize" the amp considerably. A normal power supply and amp can be designed to handle short-duration (a few ms) demands via very large oversized power supply caps, and reproduce a short dynamic peak that is 100 times its continuous rating or much more (the opposite of power supply sag). A SMPS with normal small caps sized as high-frequency ripple filter and/or a class "D" amp may have wonderful continuous rating and sound great up to that point, but no additional dynamic headroom at all, and require input compression to avoid some really bad sounds. For playing punchy bass or clean guitar, when I use a class "D" or SMPS amp I require several times more continuous power rating in order to always operate within its normal dynamic range, in which case it works very well for a clean sound. For instance, for a clean punchy bass gig I might take two Peavey classic 120 power amps with 4 6L6 in each, for a total of 120 watts continuous clean (though Peavey claims that for each), or two Peavey Deca 1200 SMPS Class D power amps, for a total of 2400 watts (no need to say "continuous" because the continuous and peak ratings are the same). Size and weght is similar. The tube amps can haandle short peak demands or be pushed beyond their limits and sound very musical. The SMPS/"D" amps must always remain within their operating limits and have their limiter switches turned on.
On the other hand, they've started making some very compact very clean-sounding "boutique" pedal steel amps using very high-powered SMPS class "D" amps built into a combo-size speaker box.
All I'm trying to say it that making a guitar amp you or someone else will love must have equal parts art and science for mutual support.
Last edited:
Greetings All,
Here is the link to the website: http://bbs.dianyuan.com/bbs/u/47/1164362877.xls
If unsuccessful, google 1164362877.xls.
Serge
Here is the link to the website: http://bbs.dianyuan.com/bbs/u/47/1164362877.xls
If unsuccessful, google 1164362877.xls.
Serge
I am currently working on a PIC based LLC SMPS.
To get around the resonant frequency problem I built up the transformer first.
I shorted out the secondaries and drove the LLC through a 100R resistor and ramped the frequency up and down to find the resonant frequency.
It turned out to be 62KHz which the PIC can provide easily. The offload frequency is 124KHz. I use an opto as feedback from the secondary smoothed voltage. I am waiting for a pcb so I cant give any more info at this stage.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.