SMPS & SS amp
I am looking for some advice in dealing with problem I have encountered with using a SMPS and a Toshiba SC-335 amplifier. Stock amplifier is rated at 190 watt AC input (117Vac) and 43 watt @ 8 ohm output. Schematics call for +/- 45Vdc and I measured +/- 42Vdc at idle off of the power supply.
I tried using a pair of Lambda SMPS rated at +/- 48Vdc at 2 amps one power supply per channel. Can adjust power supply from about 42 to 49Vdc at idle. Each output from the power supply has 10,000uF @ 63V bypassed with 0.1uf PP cap. Although I was told that a SMPS voltage does nor vary under load I can distinctly measure PS pull done at higher outputs but I can live with this just wondering. Should have lots of reserve as original unit was 190 watt and each PS is rated at 200 watts so I have doubled what I started with. I have upgraded all resistors and capacitor in signal path with Black Gates, Rikkens and some AN Tants in the amplifier.
What I have noted is that the HF is on the bright side due to the switching mode power supplies. Using a pair of stock transformer with one bridge per channel and the same 10,000uf caps with 0.1uF PP caps the HF sounds correct and normal.
Thought of trying some 0.01 or 0.001uF as well as the 0.1uF bypass caps on the 10,000uF caps. Also have read that I should have some big caps in front of the SMPS unit. So what size should be placed in front of the SMPS units as well as the voltage rating, this could be costly. Do have schematics for power supply but do not have them with me at this time and no I cannot remember the model either. Looking for all and any advise to solve this problem as I would like to use the SMPS units as I got a good deal on them. Also I am willing to open and modify the power supply just need to know what to do.
I am not imaging this as I have two modified amplifier boards with the same parts installed in both. One board is still in its original case while the second amplifier is just the parts no case. Same problem not matter which amplifier the SMPS units are hooked up to.
Hopefully we can figure this out because those little 43 watt amplifier sure sound good tweaked up, my little dragon slayers.
Thanks in advance for all your help and advise.
Happy Ears
I am looking for some advice in dealing with problem I have encountered with using a SMPS and a Toshiba SC-335 amplifier. Stock amplifier is rated at 190 watt AC input (117Vac) and 43 watt @ 8 ohm output. Schematics call for +/- 45Vdc and I measured +/- 42Vdc at idle off of the power supply.
I tried using a pair of Lambda SMPS rated at +/- 48Vdc at 2 amps one power supply per channel. Can adjust power supply from about 42 to 49Vdc at idle. Each output from the power supply has 10,000uF @ 63V bypassed with 0.1uf PP cap. Although I was told that a SMPS voltage does nor vary under load I can distinctly measure PS pull done at higher outputs but I can live with this just wondering. Should have lots of reserve as original unit was 190 watt and each PS is rated at 200 watts so I have doubled what I started with. I have upgraded all resistors and capacitor in signal path with Black Gates, Rikkens and some AN Tants in the amplifier.
What I have noted is that the HF is on the bright side due to the switching mode power supplies. Using a pair of stock transformer with one bridge per channel and the same 10,000uf caps with 0.1uF PP caps the HF sounds correct and normal.
Thought of trying some 0.01 or 0.001uF as well as the 0.1uF bypass caps on the 10,000uF caps. Also have read that I should have some big caps in front of the SMPS unit. So what size should be placed in front of the SMPS units as well as the voltage rating, this could be costly. Do have schematics for power supply but do not have them with me at this time and no I cannot remember the model either. Looking for all and any advise to solve this problem as I would like to use the SMPS units as I got a good deal on them. Also I am willing to open and modify the power supply just need to know what to do.
I am not imaging this as I have two modified amplifier boards with the same parts installed in both. One board is still in its original case while the second amplifier is just the parts no case. Same problem not matter which amplifier the SMPS units are hooked up to.
Hopefully we can figure this out because those little 43 watt amplifier sure sound good tweaked up, my little dragon slayers.
Thanks in advance for all your help and advise.
Happy Ears
Hello, I found this article. The schematic is towards the end.
http://www.convectron.com/smps/LambdaSMPS.htm
http://www.convectron.com/smps/LambdaSMPS.htm
Even more confused
Thanks alleycat
1. So pull down also occurs with SMPS underload as measured. Is there any advantage to stiffing the power supply as indicated in the article? Is this power supply acceptable for what I am using it for?
2. At 2 amps I would require about 1000uF with bypass caps as indicated. I have tried Muse at 4700uf & 10,000uF at 63Vdc with 0.1uF PP bypass caps per each voltage rail for a total of four main caps. Also tried Jenson 15,000uF @ 63Vdc with bypass caps. I found the Jenson's had the least amount of HF being to bright followed by the Muse at 10,000uF.
3. Doing research some people say that I should use less while others say it requires lots more. Also found articles that stated SMPS are superior to old iron and bridges. The only complaint I have so far is the HF is too bright regardless of what speakers or CD player I use. But again I am not suppose to hear this. What have other people experienced?
4. Tested with the following speakers Nuance Magic Ones not very good to be honest. Upgraded crossover in Athena AS-B1 big improvement over Nuance, some Polk bookshelf speakers unknown model not to bad and a pair of Okara II by North creek music.
5. Cd players used Marantz 5 disk, Arcom GCD 600 & Ah Tjoeb 4000 with upsampler. Preamp Bryston BR25.
Thanks for any and all advise that people have to offer
Thanks alleycat
1. So pull down also occurs with SMPS underload as measured. Is there any advantage to stiffing the power supply as indicated in the article? Is this power supply acceptable for what I am using it for?
2. At 2 amps I would require about 1000uF with bypass caps as indicated. I have tried Muse at 4700uf & 10,000uF at 63Vdc with 0.1uF PP bypass caps per each voltage rail for a total of four main caps. Also tried Jenson 15,000uF @ 63Vdc with bypass caps. I found the Jenson's had the least amount of HF being to bright followed by the Muse at 10,000uF.
3. Doing research some people say that I should use less while others say it requires lots more. Also found articles that stated SMPS are superior to old iron and bridges. The only complaint I have so far is the HF is too bright regardless of what speakers or CD player I use. But again I am not suppose to hear this. What have other people experienced?
4. Tested with the following speakers Nuance Magic Ones not very good to be honest. Upgraded crossover in Athena AS-B1 big improvement over Nuance, some Polk bookshelf speakers unknown model not to bad and a pair of Okara II by North creek music.
5. Cd players used Marantz 5 disk, Arcom GCD 600 & Ah Tjoeb 4000 with upsampler. Preamp Bryston BR25.
Thanks for any and all advise that people have to offer
You might want to try using a linear post-regulator on each SMPS output rail. Use the adjustable types, so you can add bypass caps to the adjust pins, to get the lowest noise and ripple on the output. See the datasheets for the LM317/LM337 and the LM338, for circuit examples etc, at national.com.
You will probably also want to try CLC "Pi"-layout filters between the SMPS and the post-regulators (since linear regs don't do much to stop high-frequency noise or ripple), by adding a series inductor after each of your SMPS output capacitors, followed by another C to ground. The J W Miller high-current toroidal inductors at mouser.com and digikey.com are cheap and will work well for this. You could start by trying something like 10 uH followed by a 2200uF or 3300 uF low-ESR cap (Mouser carries the Nichicon UHE series that would be good for this application.).
Then you can experiment with different sizes and types of caps after the regulators, and on their adjust pins.
Using both a CLC Pi filter and a linear post-regulator on each SMPS output should give you extremely-clean power rails for the amp (maybe just tens of microvolts or less peak-to-peak ripple from any SMPS high-frequency spikes), which should eliminate any sonic problems that might have been caused by the SMPS.
But you will need to make sure that the voltage at the linear regulators' inputs can stay high-enough above their output voltages for them to stay in regulation (i.e. not "drop out"). So you might also want to consider using low-dropout (LDO) adjustable regulators. Linear.com has the LT1084, LT1085, and LT1086 family, which are very good. There are also cheaper versions of them, from other companies. I've used LD1084 from mouser.com with good results. (Note that three-terminal linear regs can be used at high voltages, as long as the *difference* between their input and output voltages always stays below their "absolute maximum" rated voltage.)
If you don't like the idea of using linear regulators after the SMPS, you could try it with only the CLC filters, first.
- Tom Gootee
http://www.fullnet.com/~tomg/index.html
You will probably also want to try CLC "Pi"-layout filters between the SMPS and the post-regulators (since linear regs don't do much to stop high-frequency noise or ripple), by adding a series inductor after each of your SMPS output capacitors, followed by another C to ground. The J W Miller high-current toroidal inductors at mouser.com and digikey.com are cheap and will work well for this. You could start by trying something like 10 uH followed by a 2200uF or 3300 uF low-ESR cap (Mouser carries the Nichicon UHE series that would be good for this application.).
Then you can experiment with different sizes and types of caps after the regulators, and on their adjust pins.
Using both a CLC Pi filter and a linear post-regulator on each SMPS output should give you extremely-clean power rails for the amp (maybe just tens of microvolts or less peak-to-peak ripple from any SMPS high-frequency spikes), which should eliminate any sonic problems that might have been caused by the SMPS.
But you will need to make sure that the voltage at the linear regulators' inputs can stay high-enough above their output voltages for them to stay in regulation (i.e. not "drop out"). So you might also want to consider using low-dropout (LDO) adjustable regulators. Linear.com has the LT1084, LT1085, and LT1086 family, which are very good. There are also cheaper versions of them, from other companies. I've used LD1084 from mouser.com with good results. (Note that three-terminal linear regs can be used at high voltages, as long as the *difference* between their input and output voltages always stays below their "absolute maximum" rated voltage.)
If you don't like the idea of using linear regulators after the SMPS, you could try it with only the CLC filters, first.
- Tom Gootee
http://www.fullnet.com/~tomg/index.html
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.