WIKI-page
Hi guy's,
I have set up a basic WIKI-page for the GP.
It's located at:
Dutch Parts Group Purchases WIKI
It's far from complete yet, please ammend, correct and complete!
Have a look and don't forget to state your interests in the different parts in the table at the bottom.
Hi guy's,
I have set up a basic WIKI-page for the GP.
It's located at:
Dutch Parts Group Purchases WIKI
It's far from complete yet, please ammend, correct and complete!
Have a look and don't forget to state your interests in the different parts in the table at the bottom.
Updated the WIKI
Added an entry for power resistors and did some minor changes/corrected typo's.
Jacco,
I hope you didn't edit other things besides your entry in the inventory table because it seems we were both editing at the same time and with WIKI's things can get lost then......
I will take a look the Airlink transformers in the UK, thanks for the tip!
Added an entry for power resistors and did some minor changes/corrected typo's.
Jacco,
I hope you didn't edit other things besides your entry in the inventory table because it seems we were both editing at the same time and with WIKI's things can get lost then......
I will take a look the Airlink transformers in the UK, thanks for the tip!
psu caps
Has anyone tried Rod Elliots capacitance multplier?
I build a test setup consisting of a variac,16V transformer,35A bridge and 47000uF capacitor.
I tested with and without the multiplier and loaded the cirquit with 2.8 Amps.
Without the multiplier i got a ripple of 600mV pp.
With the multiplier the ripple was less than 1mV pp.
The losses across the transistor (2n3055) where 2.35V and stayed the same even when i doubled the load.
With a CRC or CRCRC filter the losses would be much greater and would vary with the load.
Has anyone tried this schematic out on a real amplifier and can comment on the sound?
Has anyone tried Rod Elliots capacitance multplier?
I build a test setup consisting of a variac,16V transformer,35A bridge and 47000uF capacitor.
I tested with and without the multiplier and loaded the cirquit with 2.8 Amps.
Without the multiplier i got a ripple of 600mV pp.
With the multiplier the ripple was less than 1mV pp.
The losses across the transistor (2n3055) where 2.35V and stayed the same even when i doubled the load.
With a CRC or CRCRC filter the losses would be much greater and would vary with the load.
Has anyone tried this schematic out on a real amplifier and can comment on the sound?
Capacitance multiplier
Hi kro5998,
Good results you have gotten it seems with the capacitance multiplier!
I have no experience with this circuit in an actual amp circuit but I intend to use it in an AlephX I am building at this moment as a kind of "prototype" with components I already had in my junkbox.
I want to gain some experience with the AlephX circuit, before spending big buck$$ on the realy expensive parts like high quality PSU capacitors and -transformers.
One of the things you encounter when building with parts you already have is that you will have to improvise.
In my case this means I have a number of capacitors with too low capacitance (although physicaly they are big) and higher than strictly neccesary working voltage (6800µF/63V) and a transformer with non optimal secondary voltages (12-18-24 Volts AC / 320 VA) for a "straight, C-only" PSU for a target power of abt. 100 Watts @ 8 ohms.
I figured it would be an ideal case to put the cap multiplier to work; bleeding off the excess voltage in the cap multiplier and putting it to some good use at the same time too.
I've read posts of people having good results of using this circuit in different types of amps on this forum.
I don't know about the effects on sound quality of the used type of series power transistor, but I could imagine a difference.
I remember Rod Elliot suggesting a TIP3055/TIP2955 there, but not for sound reasons.
I wonder how a MOSFET would do there soundwise?
But like I already said, no experience myself, that's what this "prototype" (for which I'm currently working on some kind of chassis to house the - rather bulky - parts) is, amongst other things, going to be used for!
Hi kro5998,
Good results you have gotten it seems with the capacitance multiplier!
I have no experience with this circuit in an actual amp circuit but I intend to use it in an AlephX I am building at this moment as a kind of "prototype" with components I already had in my junkbox.
I want to gain some experience with the AlephX circuit, before spending big buck$$ on the realy expensive parts like high quality PSU capacitors and -transformers.
One of the things you encounter when building with parts you already have is that you will have to improvise.
In my case this means I have a number of capacitors with too low capacitance (although physicaly they are big) and higher than strictly neccesary working voltage (6800µF/63V) and a transformer with non optimal secondary voltages (12-18-24 Volts AC / 320 VA) for a "straight, C-only" PSU for a target power of abt. 100 Watts @ 8 ohms.
I figured it would be an ideal case to put the cap multiplier to work; bleeding off the excess voltage in the cap multiplier and putting it to some good use at the same time too.
I've read posts of people having good results of using this circuit in different types of amps on this forum.
I don't know about the effects on sound quality of the used type of series power transistor, but I could imagine a difference.
I remember Rod Elliot suggesting a TIP3055/TIP2955 there, but not for sound reasons.
I wonder how a MOSFET would do there soundwise?
But like I already said, no experience myself, that's what this "prototype" (for which I'm currently working on some kind of chassis to house the - rather bulky - parts) is, amongst other things, going to be used for!
Capacitance multiplier reads to me as a regulated powersupply.
Mr Elliot uses 2955/3055's all the time because they are cheap
and have a prooven history, nothing wrong with that.
Audiophile did an amp design with a regulated powersupply for the output stage a number of years ago.
One of the American names produced a fully regulated poweramp once too, my memory fails me for the model and brand.
I bought the old stock of the company that produced the sound equipment for the Rijkspolitie Porsches.
TIP's were used for the output stage, i employed a number of the complete output modules as shown on the picture for BMM's
2-30volt/15 amp lab power supply.
I still have loads of these modules and chassis, already thought about using them to build something like Elliot's CM.
Voltage regulation is common practice in pre's and front end circuits, if a CM/RPS is fast and able to deliver multible amps of current i see no reason why it should not be better soundwise.
The whole idea of a powersupply for output stages is that it should be able to keep voltage level as constant as possible while delivering high currents.
That is what the big tories and large capies are there for.
I tried car batteries for the power supply of a class A power amplifier once, combined with a batteryfed preamp.
It sounded way better than the 300VA toroid/ 75.000uF powersupply.
Downside was the stack of 12 car batteries,and replacing $1000 of maintenance-free ACDelco's every 5 years.
I have seen a design that used MJ1500**'s instead of TIP's to lower the number of devices for the power regulation.
Sonically i see no reason to use other than 2955/3055's as the frequency is merely 100Hz and DC is supposed to come out, Elliot's choice for them in this case is certainly the best.
Excellent subwoofer amplifiers can be built with 2955/3055's.
If lowering transformer voltage is needed, and the powerloss is no problem Elliot's CM is a very nice idea.
You may like to read Audiophile's article on
"L'Amplificateur Fou, Fou" by Hephaistos, summer 1990.
It uses a triple V-Mos regulation for the output stage.
`
Mr Elliot uses 2955/3055's all the time because they are cheap
and have a prooven history, nothing wrong with that.
Audiophile did an amp design with a regulated powersupply for the output stage a number of years ago.
One of the American names produced a fully regulated poweramp once too, my memory fails me for the model and brand.
I bought the old stock of the company that produced the sound equipment for the Rijkspolitie Porsches.
TIP's were used for the output stage, i employed a number of the complete output modules as shown on the picture for BMM's
2-30volt/15 amp lab power supply.
I still have loads of these modules and chassis, already thought about using them to build something like Elliot's CM.
Voltage regulation is common practice in pre's and front end circuits, if a CM/RPS is fast and able to deliver multible amps of current i see no reason why it should not be better soundwise.
The whole idea of a powersupply for output stages is that it should be able to keep voltage level as constant as possible while delivering high currents.
That is what the big tories and large capies are there for.
I tried car batteries for the power supply of a class A power amplifier once, combined with a batteryfed preamp.
It sounded way better than the 300VA toroid/ 75.000uF powersupply.
Downside was the stack of 12 car batteries,and replacing $1000 of maintenance-free ACDelco's every 5 years.
I have seen a design that used MJ1500**'s instead of TIP's to lower the number of devices for the power regulation.
Sonically i see no reason to use other than 2955/3055's as the frequency is merely 100Hz and DC is supposed to come out, Elliot's choice for them in this case is certainly the best.
Excellent subwoofer amplifiers can be built with 2955/3055's.
If lowering transformer voltage is needed, and the powerloss is no problem Elliot's CM is a very nice idea.
You may like to read Audiophile's article on
"L'Amplificateur Fou, Fou" by Hephaistos, summer 1990.
It uses a triple V-Mos regulation for the output stage.
`
Attachments
Hi Jacco,
I think what it all comes down to with power supplies for power amplifiers is the capability to deliver the power quantities required by the output stages at any time, fast and without sagging in voltage or current.
For this you need a big power "reservoir".
The solution with a stack of car batteries as the power reservoir is I think almost perfect in this sense, the Nr. 1 solution, but at the cost of a high price tag and beiing a bulky solution.
The regulated power supply with big, high power, high quality transformers and big, high capacitance, high quality capacitors as the power reservoir is I think second best, but with more complexity and power dissipation losses.
The unregulated PSU with a CRC(RC) or CLC(LC) is a relatively simple solution, but sill requiring high quality transformers and capacitors as the reservoir.
The capacitance multiplier is I think a good compromise between a regulated and an unregulated PSU.
It will have much better ripple rejection than the unregulated PSU and less power dissipation than the regulated PSU but at the cost of having worse absolute voltage regulation.
I like to think of it as a CLC PSU, the "L", the inductor, beiing formed by the capacitance multiplier ("gyrator-"?) circuit.
(a gyrator is an electronic circuit without inductive components, just R's, C's and active components, emulating a (large-) inductor, for those who don't know)
A PSU with just an enormous amount of capacitance as a reservoir to suppres ripple is the simplest and most straight forward solution but will require lots of expensive capacitors and special care with respect to the high inrush currents due to the high capacitance values.
Take your pick....
I think what it all comes down to with power supplies for power amplifiers is the capability to deliver the power quantities required by the output stages at any time, fast and without sagging in voltage or current.
For this you need a big power "reservoir".
The solution with a stack of car batteries as the power reservoir is I think almost perfect in this sense, the Nr. 1 solution, but at the cost of a high price tag and beiing a bulky solution.
The regulated power supply with big, high power, high quality transformers and big, high capacitance, high quality capacitors as the power reservoir is I think second best, but with more complexity and power dissipation losses.
The unregulated PSU with a CRC(RC) or CLC(LC) is a relatively simple solution, but sill requiring high quality transformers and capacitors as the reservoir.
The capacitance multiplier is I think a good compromise between a regulated and an unregulated PSU.
It will have much better ripple rejection than the unregulated PSU and less power dissipation than the regulated PSU but at the cost of having worse absolute voltage regulation.
I like to think of it as a CLC PSU, the "L", the inductor, beiing formed by the capacitance multiplier ("gyrator-"?) circuit.
(a gyrator is an electronic circuit without inductive components, just R's, C's and active components, emulating a (large-) inductor, for those who don't know)
A PSU with just an enormous amount of capacitance as a reservoir to suppres ripple is the simplest and most straight forward solution but will require lots of expensive capacitors and special care with respect to the high inrush currents due to the high capacitance values.
Take your pick....
PSU
As mentioned on Elliots site the multiplier is not a regulated PSU.
Even with the currents i am planning to use on a low power Aleph (4 amp total per channel) it still needs a huge CRC or CLC filter to get this low a ripple current.These would require big R's or L's
which have considerably more voltage drop across them.On top of that these losses would increase or decrease with the load and thus the output voltage would vary accordingly.Try doubling the load on a CRC filter in PSU designer and see what happens to the ouput voltage.
Correct me if i am wrong but in my planned Aleph the ideal current distribution would be 50% per rail.Under load this could swing to max. 100% on one rail and 0% on the other.This means double the calculated rail current.With a CRC or CLC filter this would result in a considerably voltage drop on the 100% rail.
As mentioned on Elliots site the multiplier is not a regulated PSU.
Even with the currents i am planning to use on a low power Aleph (4 amp total per channel) it still needs a huge CRC or CLC filter to get this low a ripple current.These would require big R's or L's
which have considerably more voltage drop across them.On top of that these losses would increase or decrease with the load and thus the output voltage would vary accordingly.Try doubling the load on a CRC filter in PSU designer and see what happens to the ouput voltage.
Correct me if i am wrong but in my planned Aleph the ideal current distribution would be 50% per rail.Under load this could swing to max. 100% on one rail and 0% on the other.This means double the calculated rail current.With a CRC or CLC filter this would result in a considerably voltage drop on the 100% rail.
Mos fet in PSU
Hi byteboy, missed youre post.
I am also working with exisiting parts (transformer,bridge,caps,heatsinks) to try out the Aleph.
I have a preamp without balanced outputs so i would like to try out the Aleph first.
I am going to try a Mosfet in the multiplier cirquit but i think the voltage los would be greater (around 4 V).It won't need a darlington though.Also finding a good complimantairy pair could be a problem.P channel Mosfets would have different losses.
Hi byteboy, missed youre post.
I am also working with exisiting parts (transformer,bridge,caps,heatsinks) to try out the Aleph.
I have a preamp without balanced outputs so i would like to try out the Aleph first.
I am going to try a Mosfet in the multiplier cirquit but i think the voltage los would be greater (around 4 V).It won't need a darlington though.Also finding a good complimantairy pair could be a problem.P channel Mosfets would have different losses.
My idea for Elliot's CM would be to use higher value primary and secondary caps and a device that handles a higher current than the 15 amps the 2955/3055 do.
Or, placing several BD139/140- Tip2955/3055 in parallel with the highest caps Mr Elliott advises.
Though it is not a fully regulated powersupply and voltagedrop will be lower, the inrush current for high value capacitors will kill the TIP devices.
I mentioned the heatsink/TIP combo's because i have so many of them.
Placing a number of them in parallel with a bunch of 4700 uF Roederstein's gives maximum ripple reduction, high inrush capability, and i need not spend on it.
The TIP's i have already have 3 parallel 0.33 Ohm emitter resistors on each device, placing 4 of them in parallel would work i think.
Or, placing several BD139/140- Tip2955/3055 in parallel with the highest caps Mr Elliott advises.
Though it is not a fully regulated powersupply and voltagedrop will be lower, the inrush current for high value capacitors will kill the TIP devices.
I mentioned the heatsink/TIP combo's because i have so many of them.
Placing a number of them in parallel with a bunch of 4700 uF Roederstein's gives maximum ripple reduction, high inrush capability, and i need not spend on it.
The TIP's i have already have 3 parallel 0.33 Ohm emitter resistors on each device, placing 4 of them in parallel would work i think.
Hi All,
I somehow did not receive the E-mail notifications for updates on this thread, so I didn't look here for some time!
I updated the WIKI a little bit this morning.....
Seems interests are less than expected?
Where are you guys, posting in this thread about building amps, but not requiring any parts?
Please update the WIKI!!
I had a look at the website of Airlink Transformers. Nice and much informative website, but could however not find any info on earthscreened / shielded versions.
(From the WIKI it seems like I am the only one interested in high quality tranny's anyway.... )
Any news or info from Amplimo? Quotes/Prices? Shielding? Any other info?
kro 5998,
I think Rodd Eliot also described a symetrical CM PSU version with all NPN transistors? This maybe could be also done with all N-channel MOSFETS?
I don't know if I understand your question about the current distribution and I think this could be a major difference between an Aleph and an AlephX but the current distribution in an AlephX is said to be totally symetrical, ie. the + and the - currents are said to be (almost-) equal. So no problems with that there.
I somehow did not receive the E-mail notifications for updates on this thread, so I didn't look here for some time!
I updated the WIKI a little bit this morning.....
Seems interests are less than expected?
Where are you guys, posting in this thread about building amps, but not requiring any parts?
Please update the WIKI!!
I had a look at the website of Airlink Transformers. Nice and much informative website, but could however not find any info on earthscreened / shielded versions.
(From the WIKI it seems like I am the only one interested in high quality tranny's anyway....
)Any news or info from Amplimo? Quotes/Prices? Shielding? Any other info?
kro 5998,
I think Rodd Eliot also described a symetrical CM PSU version with all NPN transistors? This maybe could be also done with all N-channel MOSFETS?
I don't know if I understand your question about the current distribution and I think this could be a major difference between an Aleph and an AlephX but the current distribution in an AlephX is said to be totally symetrical, ie. the + and the - currents are said to be (almost-) equal. So no problems with that there.
Hello to everyone,
hope you`ve all had a Merry Christmas.
Unfortunately the participation in this group buy is very low up to now.
Maybe the name of the thread is missleading. In fact i`m the only one from outside the Netherlands that is involved. And this is obviously not enough.
I think that clear statements will have to be made who is welcome to join into the groupbuy and who can`t take part.
It also could be helpfull to start a new thread - "European Groupbuy" - to continue this one at least for the small parts like the MOSFET`s and power resistors that are easy to ship.
But the most important thing is to keep the thread alive to get people`s attention!
The Judge,
did you get any reply from Krummer?
illusionxx
hope you`ve all had a Merry Christmas.
Unfortunately the participation in this group buy is very low up to now.
Maybe the name of the thread is missleading. In fact i`m the only one from outside the Netherlands that is involved. And this is obviously not enough.
I think that clear statements will have to be made who is welcome to join into the groupbuy and who can`t take part.
It also could be helpfull to start a new thread - "European Groupbuy" - to continue this one at least for the small parts like the MOSFET`s and power resistors that are easy to ship.
But the most important thing is to keep the thread alive to get people`s attention!
The Judge,
did you get any reply from Krummer?
illusionxx
Good suggestion, Illusionxx.
Maybe European can be looked at in a broad sense.
I can imagine that some of the British, Scottish and Irish fellows would like some parts from the continent.
Channel, frontier and language wise may be a burden for them.
It seems as many buy RS parts, not my fav place to shop.
I even know someone living 5 miles from the ferry terminal to England, willing to drive the heavy parts across the channel.
I read plenty of postings by members from Germany, Austria,
"Die Schweiz", France, and Belgium.
Combining those might make this a real "Group" buy.
On another note:
i was wondering what kind of power resistors you are thinking of;
MPC's, more esoteric as Caddock 930's, or some other ?
I survived the Christmas dinner, still have 4 days to type postings till New Year's fireworks.
After 25 years of retirement i will be back on the 1st in an attempt to bomb my fingers off, maybe an eye or two.
Maybe European can be looked at in a broad sense.
I can imagine that some of the British, Scottish and Irish fellows would like some parts from the continent.
Channel, frontier and language wise may be a burden for them.
It seems as many buy RS parts, not my fav place to shop.
I even know someone living 5 miles from the ferry terminal to England, willing to drive the heavy parts across the channel.
I read plenty of postings by members from Germany, Austria,
"Die Schweiz", France, and Belgium.
Combining those might make this a real "Group" buy.
On another note:
i was wondering what kind of power resistors you are thinking of;
MPC's, more esoteric as Caddock 930's, or some other ?
I survived the Christmas dinner, still have 4 days to type postings till New Year's fireworks.
After 25 years of retirement i will be back on the 1st in an attempt to bomb my fingers off, maybe an eye or two.
I haven`t thought about the power R`s yet and i even don`t know which ones would be recommendable - most probably low induction types. With large quantities we could be able to get the esoteric parts for a reasonable price.
Maybe some people are also interested in highest quality caps for the small values in the circuit (220 uF).
Everybody`s recommendations are welcome.
To all the people that are new to this thread:
There is a WIKI to see what quantities a group buy can reach, that can be found here WIKI
So please sign in if you are interested or post if there are any questions.
illusionxx
Maybe some people are also interested in highest quality caps for the small values in the circuit (220 uF).
Everybody`s recommendations are welcome.
To all the people that are new to this thread:
There is a WIKI to see what quantities a group buy can reach, that can be found here WIKI
So please sign in if you are interested or post if there are any questions.
illusionxx
I'm sorry I'm not in the market for an Aleph-X rightnow (just contributed some hints). Currently I'm trying to finish my Aleph5 (for which all parts are already acquired) and a LCaudio Zappulse (for the lows). I purchased 4 Rifa's for the latter two weeks ago (very nice indeed).
I think for a succesfull groupbuy you have to a better focus what its for, so you have a good chance of getting a good discount. Having a single vendor helps with this of course. Also you need sombody who takes the lead (and who has enough time to do the organizing).
Personally at the momemt I'm only interested in some diodes such as scottkys or good bridge types such as Ixys for a 45 VAC PS
I think for a succesfull groupbuy you have to a better focus what its for, so you have a good chance of getting a good discount. Having a single vendor helps with this of course. Also you need sombody who takes the lead (and who has enough time to do the organizing).
Personally at the momemt I'm only interested in some diodes such as scottkys or good bridge types such as Ixys for a 45 VAC PS
current distribution
Hi byteboy, your'e right about the current distribution.Because the Aleph X is basically 2 bridged aleph's with inverted inputs the load differences cancel each other out.
I tried a N-channel mosfet in the Capacitance multiplier but the losses (drain-source voltage) where greater compaired to using a 2n3055.
I plan using the pair MJ11015/MJ11016 which are darlingtons rated at 30A.The Inrush current of the capacitors after the transistor are limited by the chargetime of the capacitor at the base of the transistor.
Do you have a link to the schematic on Elliots site about the PSU with only NPN transistors?Then I could use the cheaper 2N3771.(1 euro a piece).
Best wishes
kro5998.
Hi byteboy, your'e right about the current distribution.Because the Aleph X is basically 2 bridged aleph's with inverted inputs the load differences cancel each other out.
I tried a N-channel mosfet in the Capacitance multiplier but the losses (drain-source voltage) where greater compaired to using a 2n3055.
I plan using the pair MJ11015/MJ11016 which are darlingtons rated at 30A.The Inrush current of the capacitors after the transistor are limited by the chargetime of the capacitor at the base of the transistor.
Do you have a link to the schematic on Elliots site about the PSU with only NPN transistors?Then I could use the cheaper 2N3771.(1 euro a piece).
Best wishes
kro5998.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.