Please can we explode two myths here, from Audio Folklore..........
1. SMPS's are no good for powering audio amplifiers
2. Class D amplifiers can't handle input rail voltage sag.
******************************************
1. Surley cannot be true....usual reason is said to be noise from the SMPS...but good layout and decoupling and filtering can solve that.
2. DSP's are the masters of all filtration and sound processing....you can definetely use a DSP to correct for rail input voltage sag on a class d amps input rails.
1. SMPS's are no good for powering audio amplifiers
2. Class D amplifiers can't handle input rail voltage sag.
******************************************
1. Surley cannot be true....usual reason is said to be noise from the SMPS...but good layout and decoupling and filtering can solve that.
2. DSP's are the masters of all filtration and sound processing....you can definetely use a DSP to correct for rail input voltage sag on a class d amps input rails.
Yes they can be, but Bass pumping is a problem, I used a resonate switch mode, with a dumping design to reduce this effects.
It is interesting that you put these #1 and #2 statements in the same topic because they are at conflict with each other.
If the intention is powering a class D amp and input rail voltage sag is tolerable (which it usually is unless the amp gain vs amp input voltage is excessive for the design), you just removed the main strength of the SMPS. Otherwise the way you worded #1 was complimentary to furthering an argument rather than the more neutral way of stating things which would be:
"SMPS have strengths and weaknesses for powering audio amplifiers."
Certainly their efficiency is a strength, and some feel the high frequency noise makes music "livelier" (better sounding), and for portable audio they can be lighter weight, but if you are powering an amp the high frequency noise of a SMPS is generally undesirable.
You can lay it out good, filter it good, and what do you have? Additional time, money, space spent in trying to achieve the qualities of a linear PSU, but with shorter expected lifespan, more time spent troubleshooting when it fails someday.
Yes you can go to lengths to analyze, measure, test... make SMPS output have very low ripple, but towards what end? A significant additional effort should bring significant additional benefit, not just reinventing the wheel.
If the intention is powering a class D amp and input rail voltage sag is tolerable (which it usually is unless the amp gain vs amp input voltage is excessive for the design), you just removed the main strength of the SMPS. Otherwise the way you worded #1 was complimentary to furthering an argument rather than the more neutral way of stating things which would be:
"SMPS have strengths and weaknesses for powering audio amplifiers."
Certainly their efficiency is a strength, and some feel the high frequency noise makes music "livelier" (better sounding), and for portable audio they can be lighter weight, but if you are powering an amp the high frequency noise of a SMPS is generally undesirable.
You can lay it out good, filter it good, and what do you have? Additional time, money, space spent in trying to achieve the qualities of a linear PSU, but with shorter expected lifespan, more time spent troubleshooting when it fails someday.
Yes you can go to lengths to analyze, measure, test... make SMPS output have very low ripple, but towards what end? A significant additional effort should bring significant additional benefit, not just reinventing the wheel.
Last edited:
sorry i am not arguing here, more just grinding out my point.....sory to rant on..
100W worth of output power from a mains transformer will weigh about 2Kg.
thats way too much weight to have in a product and have to ship that product around the world in any kind of quantity
Also, supposing you change your product a bit......say you want to change your vout from +-50v to +-40v......
-well thats easy with an smps, because you can quickly change course with a ferrite transformer, and start getting new stock quickly wound up.........and cheaply wound up.
...wheras if you were using mains transformers, you'd have to get a big back-stock pre-wound, so you'd be chucking the back-stock away....very expensive........
......any small backstocks of ferrite transformers could be chucked away and it'd be less financial loss.
you can even wind your own ferrite transformer, and use it straight away on your pcb.....you cannot even dream of winding your own mains transformer.......not without huge financial outlay.
I am always hearing how hard SMPS's are, and how complicated they are to develop.
-I've seen electronics assemblers , who left school at 15 with few qualifications, step up and become valuable SMPS design engineers at the same company.
100W worth of output power from a mains transformer will weigh about 2Kg.
thats way too much weight to have in a product and have to ship that product around the world in any kind of quantity
Also, supposing you change your product a bit......say you want to change your vout from +-50v to +-40v......
-well thats easy with an smps, because you can quickly change course with a ferrite transformer, and start getting new stock quickly wound up.........and cheaply wound up.
...wheras if you were using mains transformers, you'd have to get a big back-stock pre-wound, so you'd be chucking the back-stock away....very expensive........
......any small backstocks of ferrite transformers could be chucked away and it'd be less financial loss.
you can even wind your own ferrite transformer, and use it straight away on your pcb.....you cannot even dream of winding your own mains transformer.......not without huge financial outlay.
I am always hearing how hard SMPS's are, and how complicated they are to develop.
-I've seen electronics assemblers , who left school at 15 with few qualifications, step up and become valuable SMPS design engineers at the same company.
Lots of things that weigh more than 2kg are shipped world wide, daily. It doesn't seem to be an issue to me more than a slight additional shipping cost. Contrast a SMPS, a much larger # of parts has to be shipped from several different locations in order to build them.
If you change your product you would be sending the SMPS back to the manufacturer, not opening and modifying it yourself as then the warranty is gone and it is no longer approved by any safety agency when altered.
I don't know what you mean by "chucked away". Any good that has value is not just thrown in the trash, they would be resold and the same is true of the wrong order for a SMPS with a voltage you don't use.
I don't feel that being able to wind your own transformer is significant, not when there are transformers available for purchase. However since the scenario posed was changing voltage output from +-50V to +-40V, I feel it is easier to take some windings off a transformer than to desolder one and pull it out of a SMPS, then do the same or rewind it, then reverse engineer the feedback loop enough to modify it so it regulates for the new voltage.
Anyone can develop a SMPS, but let's look at the ones that exist as products for sale... in most cases, one with exceptionally clean output costs more, not less than a linear PSU does because the majority of SMPS are designed for equipment that does not need ultra clean power. Further, to realistically expect that an alteration to a SMPS really cleans up the output enough, a scope is needed to assess the ripple and that is an expensive piece of equipment that many DIY'ers do not own.
For the DIY amp builder, even if they custom design their own from scratch to get cleaner power, all the individual parts to make it, bought in low quantities will cost a great deal of money. We could say that maybe they will find some discounted sale price on surplus SMPS stock and alter it to their needs, but then we would have to consider finding some discount sale price on a surplus linear PSU's components too.
If you change your product you would be sending the SMPS back to the manufacturer, not opening and modifying it yourself as then the warranty is gone and it is no longer approved by any safety agency when altered.
I don't know what you mean by "chucked away". Any good that has value is not just thrown in the trash, they would be resold and the same is true of the wrong order for a SMPS with a voltage you don't use.
I don't feel that being able to wind your own transformer is significant, not when there are transformers available for purchase. However since the scenario posed was changing voltage output from +-50V to +-40V, I feel it is easier to take some windings off a transformer than to desolder one and pull it out of a SMPS, then do the same or rewind it, then reverse engineer the feedback loop enough to modify it so it regulates for the new voltage.
Anyone can develop a SMPS, but let's look at the ones that exist as products for sale... in most cases, one with exceptionally clean output costs more, not less than a linear PSU does because the majority of SMPS are designed for equipment that does not need ultra clean power. Further, to realistically expect that an alteration to a SMPS really cleans up the output enough, a scope is needed to assess the ripple and that is an expensive piece of equipment that many DIY'ers do not own.
For the DIY amp builder, even if they custom design their own from scratch to get cleaner power, all the individual parts to make it, bought in low quantities will cost a great deal of money. We could say that maybe they will find some discounted sale price on surplus SMPS stock and alter it to their needs, but then we would have to consider finding some discount sale price on a surplus linear PSU's components too.
Last edited:
With a mains transformer, you're at about 2.7Kgs per 100W.
...So thats 7.1Kg for 300W.
...Thats just too heavy, freight carriers are charging people for weight now, as opposed to volume, like before..
..in a shop, a customer with money to spend wont buy the heavier unit containing the Mains transformer....if a much lighter unit does the same job.
once youve got Far Eastern manufacturing going, you can get an SMPS cheaply made......at 300W, i reckon the cost of mains transformers is more than SMPS cost.
you take your guitar amplifier prototype build to Harrods or somewhere and suggest them sell it as a new line.......then there marketing manager trys to pick it up and cant lift it...straight away they've got problems...... are there staff going to want to lug these things around.?................
...So thats 7.1Kg for 300W.
...Thats just too heavy, freight carriers are charging people for weight now, as opposed to volume, like before..
..in a shop, a customer with money to spend wont buy the heavier unit containing the Mains transformer....if a much lighter unit does the same job.
once youve got Far Eastern manufacturing going, you can get an SMPS cheaply made......at 300W, i reckon the cost of mains transformers is more than SMPS cost.
you take your guitar amplifier prototype build to Harrods or somewhere and suggest them sell it as a new line.......then there marketing manager trys to pick it up and cant lift it...straight away they've got problems...... are there staff going to want to lug these things around.?................
Actually, it costs me about $10 to have a 300VA transformer shipped. If I had a SMPS shipped it would cost at least $6 and usually closer to the same $10. The key thing to remember is for small items there is an initial charge to ship anything in a box but the cost does not go up much unless it is extremely heavy and a single transformer is not extremely heavy.
It is irrelevant though, that a "good" SMPS, delivered, will cost more in total including shipping. The costs may differ in different, more remote areas of the world but in areas with reasonable population density nearby, there will be distribution warehouses so things like SMPS or transformers are shipped there in bulk, unless it is a relatively rare item but transformers and SMPS aren't particularly rare.
If we want to consider cost we should also consider cost per year. The SMPS may have only half the lifespan, and the time spent diagnosing a fault and hunting down a repair part is higher. Time has to be considered of some monetary value to mortal beings.
In a shop, the last thing a customer is usually thinking is what does this amp weigh - but actually, I usually pick the heavier item if I can discern no other differences because generally speaking, heavier goods are usually higher quality, that a manufacturer did not skimp on metal in the chassis or heatsinks, or other structural materials or components is at least a minimal level of assurance they didn't try to cut costs as much as possible - the same goes for SMPS, a lighter weight SMPS is generally an inferior SMPS (to a heavier SMPS).
I can assure you the SMPS costs more here in the USA. I mean something with clean output not just a low end junk SMPS.
I have no idea where you are getting the idea that amps people have built for dozens of years are suddenly too heavy to lift. It is not a factor except with specialized portable gear - gear that is to be used while it is being carried around continually and such gear tends to run from batter power - but in that case I am all for a buck or boost switching circuit to arrive at the right operating voltage.
It is irrelevant though, that a "good" SMPS, delivered, will cost more in total including shipping. The costs may differ in different, more remote areas of the world but in areas with reasonable population density nearby, there will be distribution warehouses so things like SMPS or transformers are shipped there in bulk, unless it is a relatively rare item but transformers and SMPS aren't particularly rare.
If we want to consider cost we should also consider cost per year. The SMPS may have only half the lifespan, and the time spent diagnosing a fault and hunting down a repair part is higher. Time has to be considered of some monetary value to mortal beings.
In a shop, the last thing a customer is usually thinking is what does this amp weigh - but actually, I usually pick the heavier item if I can discern no other differences because generally speaking, heavier goods are usually higher quality, that a manufacturer did not skimp on metal in the chassis or heatsinks, or other structural materials or components is at least a minimal level of assurance they didn't try to cut costs as much as possible - the same goes for SMPS, a lighter weight SMPS is generally an inferior SMPS (to a heavier SMPS).
I can assure you the SMPS costs more here in the USA. I mean something with clean output not just a low end junk SMPS.
I have no idea where you are getting the idea that amps people have built for dozens of years are suddenly too heavy to lift. It is not a factor except with specialized portable gear - gear that is to be used while it is being carried around continually and such gear tends to run from batter power - but in that case I am all for a buck or boost switching circuit to arrive at the right operating voltage.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.