I was hoping for some input / suggestions...
My design consists of the following:
1. I am making an educated guess and assuming that I will have 13.8V rails, all said and done. I plan on playing with inductance / series resistance in the pi filter to get somewhere near that value. I will be using 105-120 µF per rail filtering, probably with a 2mH- .30 ohm DC resistance choke seperating two equal banks of caps... any suggetions?
2. My speakers are fairly low impedance- 3.2 ohms minimum. I would like the amps to be able to deliver into a fairly low impedance without excessive current limiting. Considering my available heat sinking and rail voltages, I am expecting about 60W into 4 ohms, current limited.
3. I have about 240 Watts of heatsinking per channel.
4. I am going to use either (A) 6 output devices per side with .47 ohm source resistors for about 3.3A per side of idle current, or (B) 8 output devices per side with .6 ohm source resistors for about 3.5A per side idle current. I am guessing that choice B will be a little much for my heat sinks, but with extra devices I can push them harder on a hotter heat sink. Any thoughts?
5. I don't plan on using the transistors that limit signal to the output devices. Any reason I should reconsider this? Also, if I go with choice B (see 4) I will probably use .3 ohm resistors on the sources of the lower output transistors.
6. What (cheap) resistors would be a good bet for the current sensing resistors / source resistors on the output stage? What type?
Any info / experience and suggestions are welcome.
-NS
My design consists of the following:
1. I am making an educated guess and assuming that I will have 13.8V rails, all said and done. I plan on playing with inductance / series resistance in the pi filter to get somewhere near that value. I will be using 105-120 µF per rail filtering, probably with a 2mH- .30 ohm DC resistance choke seperating two equal banks of caps... any suggetions?
2. My speakers are fairly low impedance- 3.2 ohms minimum. I would like the amps to be able to deliver into a fairly low impedance without excessive current limiting. Considering my available heat sinking and rail voltages, I am expecting about 60W into 4 ohms, current limited.
3. I have about 240 Watts of heatsinking per channel.
4. I am going to use either (A) 6 output devices per side with .47 ohm source resistors for about 3.3A per side of idle current, or (B) 8 output devices per side with .6 ohm source resistors for about 3.5A per side idle current. I am guessing that choice B will be a little much for my heat sinks, but with extra devices I can push them harder on a hotter heat sink. Any thoughts?
5. I don't plan on using the transistors that limit signal to the output devices. Any reason I should reconsider this? Also, if I go with choice B (see 4) I will probably use .3 ohm resistors on the sources of the lower output transistors.
6. What (cheap) resistors would be a good bet for the current sensing resistors / source resistors on the output stage? What type?
Any info / experience and suggestions are welcome.
-NS
1) you mean 105,000 uF? My suggestion? Play with duncan amps simulator it's super easy. It will give you enormous pleasure to see what happens with more or less caps, more or less X or R. Stop when you think you have low enough ripple or you hit your budget limit in PS parts or the best noise compromise.
2) right
3) swell
4) do you mean 6 CS and 6 gain or 3 CS and 3 gain/side/chn? I am using 1+1/side/chn with 5 amps/chn at 14-0-14 VDC and everything seems to be under control. and my heatsink is rated at +21 degC above RT @ 150W. Mosfet case are at 70 C.
I would dare say that with your heatsink capabilities you can push both the voltage and the current up slightly with just 3 parallel mosfets/"gain unit". According to my poor man's MC simulation 6 devices would be easier to match.
And if it's true what Nelson says that what matters in terms of sound is the total power dissipated you should be in good shape, it's just more money.
5) at this time I don't have them in either, but I could just as well have them. It makes it a little bit easier to tune up, not having them that is.
6) Any resistor you want but I am known not to give a **** about parts as long as they have low tolerance, good thermal specs and look good. I do use "modern" components though.
2) right
3) swell
4) do you mean 6 CS and 6 gain or 3 CS and 3 gain/side/chn? I am using 1+1/side/chn with 5 amps/chn at 14-0-14 VDC and everything seems to be under control. and my heatsink is rated at +21 degC above RT @ 150W. Mosfet case are at 70 C.
I would dare say that with your heatsink capabilities you can push both the voltage and the current up slightly with just 3 parallel mosfets/"gain unit". According to my poor man's MC simulation 6 devices would be easier to match.
And if it's true what Nelson says that what matters in terms of sound is the total power dissipated you should be in good shape, it's just more money.
5) at this time I don't have them in either, but I could just as well have them. It makes it a little bit easier to tune up, not having them that is.
6) Any resistor you want but I am known not to give a **** about parts as long as they have low tolerance, good thermal specs and look good. I do use "modern" components though.
Grataku,
Thanks for the help.
I'll give the duncanamps sim a try. Yes, I meant 105000µF (typo).
I meant 6 total gain devices + 6 total current source devices- or 3+3 per side.
I may have been a little optimistic with my heatsinking- looking at it again I can go about 200Watts with a 24 deg rise above ambient. 220Watts would be about 30 deg. I calculate that with 13.8V and 3.5A bias per side, I will be dissipating about
193 Watts at idle, which should be alright. I may try and push it a little higher.
-NS
Thanks for the help.
I'll give the duncanamps sim a try. Yes, I meant 105000µF (typo).
I meant 6 total gain devices + 6 total current source devices- or 3+3 per side.
I may have been a little optimistic with my heatsinking- looking at it again I can go about 200Watts with a 24 deg rise above ambient. 220Watts would be about 30 deg. I calculate that with 13.8V and 3.5A bias per side, I will be dissipating about
193 Watts at idle, which should be alright. I may try and push it a little higher.
-NS
a problem
If the duncanamps power supply designer is correct, my transformer will not have enough voltage! It has 13.5V secondaries, but the simulator says that if I run it at 7A total bias, I will be down to about 9 volts! Does anyone know how accurate the designer is? Do I need to go find new transformers? I was using 68mF capacitance with a 2mH choke- .3 ohms and then another 34mF capacitance.
-NS
If the duncanamps power supply designer is correct, my transformer will not have enough voltage! It has 13.5V secondaries, but the simulator says that if I run it at 7A total bias, I will be down to about 9 volts! Does anyone know how accurate the designer is? Do I need to go find new transformers? I was using 68mF capacitance with a 2mH choke- .3 ohms and then another 34mF capacitance.
-NS
Why don't we have the discussion together in the threat which has more or less the same purpose and where we have quite a bit on feedback to the configuration you have in mind:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=8933
I would be very happy to receive your feedback on the current source / schematics I posted for this configuration as well.
Best Regards
http://www.diyaudio.com/forums/showthread.php?s=&threadid=8933
I would be very happy to receive your feedback on the current source / schematics I posted for this configuration as well.
Best Regards
NS
It doesn't make sense, not with .3 ohm DCR. What kind of voltage regulation do you use on the trafo? IE what's the internal resistance of the secondary. Check the resistance of the CAPS as well, it has to be a realistically low value.
I have learned quite a few things about PS with that program.
It doesn't make sense, not with .3 ohm DCR. What kind of voltage regulation do you use on the trafo? IE what's the internal resistance of the secondary. Check the resistance of the CAPS as well, it has to be a realistically low value.
I have learned quite a few things about PS with that program.
Grataku,
Thanks again for the suggestions. I did play around with it a little more, and the ESR on the caps has a lot to do with it, just as you said. I still will be at just under 12 volts, or so says the ps designer. So, I decided to use the .6 ohm resistors and use 3+3 devices per side. If it isn't pushing enough current, I plan to leave room on the chassis for a 4th pair. So, it's sort of a compromise- voltage for current.
BTW, I was using .050ohms for the secondary of the tranformer. I believe that is around 10% regulation (correct me if I'm wrong).
Thanks.
-NS
Thanks again for the suggestions. I did play around with it a little more, and the ESR on the caps has a lot to do with it, just as you said. I still will be at just under 12 volts, or so says the ps designer. So, I decided to use the .6 ohm resistors and use 3+3 devices per side. If it isn't pushing enough current, I plan to leave room on the chassis for a 4th pair. So, it's sort of a compromise- voltage for current.
BTW, I was using .050ohms for the secondary of the tranformer. I believe that is around 10% regulation (correct me if I'm wrong).
Thanks.
-NS
NS,
the resistance of your chokes is a bit high at .3 ohms. This alone give about 2,1V loss (with 7A).
I think the internal resistance of your transformer is set a bit too high, 10-20mOhms seems to be realistic.
I get a faktor of about 1 (AC to DC voltage) with a PI-filter supply for an Aleph-X (16V)
william
the resistance of your chokes is a bit high at .3 ohms. This alone give about 2,1V loss (with 7A).
I think the internal resistance of your transformer is set a bit too high, 10-20mOhms seems to be realistic.
I get a faktor of about 1 (AC to DC voltage) with a PI-filter supply for an Aleph-X (16V)
william
William,
They were 2mH 14 ga. air core... can you suggest a more suitable part? I haven't bought them yet.
I will try the transformer at a different resistance, like you said.
What total current (both sides)are your amps idling at?
THanks for the help
-NS
NS,
they will be idling at around 10A for both halves. A more suitable part would be a torobar choke wich has around 0.12 ohm at 2.2mH
william
they will be idling at around 10A for both halves. A more suitable part would be a torobar choke wich has around 0.12 ohm at 2.2mH
william
William,
I definitely like the idea of a choke with a lower dc resistance. Is core saturation an issue when using a choke in this way?
-NS
I definitely like the idea of a choke with a lower dc resistance. Is core saturation an issue when using a choke in this way?
-NS
NS,
the specs say they saturate at 11A so that should be enough. I don´t know if it would hurt if they were only specified for 5A though.
william
the specs say they saturate at 11A so that should be enough. I don´t know if it would hurt if they were only specified for 5A though.
william
NS,
they are from Intertechnik wich is well known in Germany but probably not in the USA. Search for torobar and you´ll find a link to there homepage.
william
they are from Intertechnik wich is well known in Germany but probably not in the USA. Search for torobar and you´ll find a link to there homepage.
william
Here is the link:
http://www.intertechnik.de/cgi-loca...Elektronikbauteile&&Spulen&&Corobar&&none&&15
William,
I have been looking inti the intertechnik coils as well, but have not found any specs on the saturation current. From where did you get that from ?
Best Regards
http://www.intertechnik.de/cgi-loca...Elektronikbauteile&&Spulen&&Corobar&&none&&15
William,
I have been looking inti the intertechnik coils as well, but have not found any specs on the saturation current. From where did you get that from ?
Best Regards
Blitz,
from this site go to torobar (not corobar), klick on the 2.2mH one and somewhere on the lower left is "sät 10,7A"
william
from this site go to torobar (not corobar), klick on the 2.2mH one and somewhere on the lower left is "sät 10,7A"
william
Oops, sorry, yes, found it. I asked myself wether these coils will vibrate in a PSU or not. Maybe the choke (Trafokernspulen) which are vaccumcoated are quieter ? I asked intertechnik and wait for an answer. Anyone used the the Torobars in a PSU yet ? In a Pi-Filter or Choke-Input ?
Blitz,
there´s no reason they should vibrate. Use PSU designer and look at the current going through the coil. There´s only a small ripple without any sharp peaks.
william
there´s no reason they should vibrate. Use PSU designer and look at the current going through the coil. There´s only a small ripple without any sharp peaks.
william
Well, I was thinking about using them as well as the first coil in a choke-input-setup and than you have much more ripple...
- Status
- Not open for further replies.