in addition ? thats what Im thinking now. A-n-y b-o-d-y-o-u-t-t-h-e-r-e ?
Im presuming only a few people have made the jx...😛
Im presuming only a few people have made the jx...😛
Measure the Idss of the JFET in the CCS, with that number you'll find the corrresponding transconductance in the datasheet of the J74.
1 divided by that mS number is the apparant resistance of the JFET.
Adding a resistor value at the Source of the JFET will lower the drain current to the level you want, Id is reduced from the Idss value by the total series resistance of Rs and the virtual resistance value of the JFET.
1 divided by that mS number is the apparant resistance of the JFET.
Adding a resistor value at the Source of the JFET will lower the drain current to the level you want, Id is reduced from the Idss value by the total series resistance of Rs and the virtual resistance value of the JFET.
Bah, Cheers Jacco measured the wrong part. That'll be cause I dont know wtf Im doing. Think I know how to work it out now. Soooooooo just add a 100r to ground on the outputs and Im done I guess....
The J-X is a balanced version, your J109 is 5% Off.
No itch to fiddle wih the drain resistor values ?
No itch to fiddle wih the drain resistor values ?
Inrush Current Limiter & Capacitor Ripple Specification
With Respect To(WRT) capacitor ripple capacity, is it a specification
to maximize or is there a limit with dimminishing returns, because I was told a long time ago that it is better to have many capacitors to make up the required capacitance insted of just one or two, I dont know the logic in the statement though and it may have been marketing for a David Hafler DH100 retrofit kit from a third party, the DIY'r selling statement was those big can capacitors in the DH100 are no good compared to many smaller capacitors making up the same capacitance.
I dont know how to read the Capacitor Ripple Specification from the Data Sheets, is a higher number good or a smaller number good ? I know that all capacitors do not give out this specification, I suspect because the number is bad WRT the competition capacitors ?
So, insted of me useing three 33KmicroFerrid capacitors I should use ten 10KmicroFerrid in parallel making up the four places I will use in the 100KmicroFerrid +/-20% capacitance in the CLCRCRC Power Supply ?
WRT the inductor I found out that the 16gage Erse Super Q is 16amps I belieive but I could be wrong, it was used in the Aleph 3 amplifier succsessfully, so I will purchase the 14gage Erse Super Q someware insted of the 16gage one at Parts Express Erse Super Q 2.0mH 16 AWG 500W Inductor | Parts-Express.com Parts Express dose not stock the 14gage inductor, I believe that it should have better amps power specification, I will find this out if it is public information before purchasing.
Finally I am down to the last question for this message. Can I use just a AMETHERM MEGAsurge part#MS35-10018(see attachment Data Sheet) Ametherm's new Inrush Current Limiter insted of all of the parts on the "PowerSwitch-CurrentLimiter" schematic attached to this message ?
Thank You again for your help, all comments are usefull and I hope that some of the questions I have had solutions for helps others as well in the construction your own Aleph J-X amplifier.
With Respect To(WRT) capacitor ripple capacity, is it a specification
to maximize or is there a limit with dimminishing returns, because I was told a long time ago that it is better to have many capacitors to make up the required capacitance insted of just one or two, I dont know the logic in the statement though and it may have been marketing for a David Hafler DH100 retrofit kit from a third party, the DIY'r selling statement was those big can capacitors in the DH100 are no good compared to many smaller capacitors making up the same capacitance.
I dont know how to read the Capacitor Ripple Specification from the Data Sheets, is a higher number good or a smaller number good ? I know that all capacitors do not give out this specification, I suspect because the number is bad WRT the competition capacitors ?
So, insted of me useing three 33KmicroFerrid capacitors I should use ten 10KmicroFerrid in parallel making up the four places I will use in the 100KmicroFerrid +/-20% capacitance in the CLCRCRC Power Supply ?
WRT the inductor I found out that the 16gage Erse Super Q is 16amps I belieive but I could be wrong, it was used in the Aleph 3 amplifier succsessfully, so I will purchase the 14gage Erse Super Q someware insted of the 16gage one at Parts Express Erse Super Q 2.0mH 16 AWG 500W Inductor | Parts-Express.com Parts Express dose not stock the 14gage inductor, I believe that it should have better amps power specification, I will find this out if it is public information before purchasing.
Finally I am down to the last question for this message. Can I use just a AMETHERM MEGAsurge part#MS35-10018(see attachment Data Sheet) Ametherm's new Inrush Current Limiter insted of all of the parts on the "PowerSwitch-CurrentLimiter" schematic attached to this message ?
Thank You again for your help, all comments are usefull and I hope that some of the questions I have had solutions for helps others as well in the construction your own Aleph J-X amplifier.
Ripple current capability must exceed the in circuit ripple current. Beyond a small margin, increasing ripple capability buys some extended lifetime. There are many ways of looking at using multiple caps. As noted earlier, in the same type you will likely have more ripple current capability with two smaller caps than a single of equal total capacity. You will also lower the effective ESR with multiple caps, as the ESR of each is in parallel. Lower ESR means increased ripple current and in the back of my head I remember reading something about potential resonances.
Where does the point of diminishing returns hit? Your call. Is the additional labor and hardware required to mount multiple caps worth the increased current capability and lower ESR?
An inductor that won't saturate its core at 16A should be fine at 10A. The 14 gauge will likely be built on the same core, the only advantage being the lower DCR.
The MegaSurge is a negative temperature coefficient thermistor, another version of the CL60 used in so many Pass amps and clones. You won't get the DC blocking shown in the switch circuit, if you think that you need it.
Where does the point of diminishing returns hit? Your call. Is the additional labor and hardware required to mount multiple caps worth the increased current capability and lower ESR?
An inductor that won't saturate its core at 16A should be fine at 10A. The 14 gauge will likely be built on the same core, the only advantage being the lower DCR.
The MegaSurge is a negative temperature coefficient thermistor, another version of the CL60 used in so many Pass amps and clones. You won't get the DC blocking shown in the switch circuit, if you think that you need it.
WRT the CL60 it is only rated to pass a max of 5Amps, which will be fine with the other amps it was used in, I felt that the 5 amps max continuas current will be to low for this project, so I chose the AMETHERM MEGAsurge #MS35-10018 because it is rated at 18amps max continuas current while rated at 25degree 10ohms, same as the CL60. Have I analysed this properly ??
the specification I need to know how to read is the Capacitor Data Sheet is the "Ripple Current Multiplier" and the "Ripple Current Ambient Temperature Factors", again are higher values or lower values better so I can do some valuable analyses between Capacitors.
I am now thinking of useing nineteen 6.8KmicroFerrid in paralell totaling 129.2KmicroFerrid for each group required by the CLCRCRC Power Supply for a total of 304 capacitors for two monoblock Aleph J-X amplifiers. What do you think, is this going to far, it will not cost much more because the price drops significantly over 300 units ordered, that is why I came up with 19 paralell capacitors per bank ?
Thank You, I am almost ready to order parts, but I am scared of making a mistake, not that the amp will not work but I may need to upgrade it next year because knowing that other parts may have been better choice, so I will try and do all of the research I can now, but I realy need to have this amp working in my stereo now because I have just finnished speakers I have been researching for years that sound amazing but I need another amp for above 100Hz, what do I do now, buy or research?
Thank You For Your Help
Last edited:
IIRC, You'll be running at around 500W, which is ~4A @120V. Your calculations for the NTC are correct as far as they go. The other side is how low the resistance drops at the operating point. The CL-60 will be within its limits but much lower hot resistance. The Mega Surge MS35 20010 might be a good choice - higher cold resistance (lower inrush current) and lower hot resistance than the -0018.
Without seeing the data sheets in question, I'll guess that those are derating factors. The smaller the number the shorter the life. There is usually an example of how to use them in the data sheet.
15 caps per position seems like an awful lot of assembly work for potentially a marginal gain. Will you make PCBs to mount them? Use heavy copper and wide traces.
It sounds like analysis paralysis is setting in. No matter how long you study, there will always be some idea that might sound better that would require a redesign to incorporate once you start your build. Build, enjoy, and if you really think you need to upgrade in a year or so, go ahead.
Edit: You can also use lower rated NTCs by using one in each leg of the transformer primary (assuming you have a dual primary transformer)
Without seeing the data sheets in question, I'll guess that those are derating factors. The smaller the number the shorter the life. There is usually an example of how to use them in the data sheet.
15 caps per position seems like an awful lot of assembly work for potentially a marginal gain. Will you make PCBs to mount them? Use heavy copper and wide traces.
It sounds like analysis paralysis is setting in. No matter how long you study, there will always be some idea that might sound better that would require a redesign to incorporate once you start your build. Build, enjoy, and if you really think you need to upgrade in a year or so, go ahead.
Edit: You can also use lower rated NTCs by using one in each leg of the transformer primary (assuming you have a dual primary transformer)
Someone asked a few posts back if Peter had built one of the XJ amps. Has anyone else got one up and running.
I purchased four of these boards several months back and have been trying to decide if just constructing the stock J would not in itself be a great amp. Nelson thought enough of this design to reengineer the earlier Aleph 3 and incorporate into the First Watt lineup.
I would like to see a progress report on the completed XJ before plunging into this with huge heatsinks and the like.
Bob, I like the analysis paralysis nice one. Most of the DIY efforts at perfection will hardly be audible anyway. The real meat in this amp is the decades of experience this designer has given us.
Tad
I purchased four of these boards several months back and have been trying to decide if just constructing the stock J would not in itself be a great amp. Nelson thought enough of this design to reengineer the earlier Aleph 3 and incorporate into the First Watt lineup.
I would like to see a progress report on the completed XJ before plunging into this with huge heatsinks and the like.
Bob, I like the analysis paralysis nice one. Most of the DIY efforts at perfection will hardly be audible anyway. The real meat in this amp is the decades of experience this designer has given us.
Tad
mounting 16 capacitors together without much silver-copper wire being used will be a problem. Maybe I could put four together forming a cross with the wire attaching the four in the middle using only about 1inch of wire, then I can attach four of the four-Capacitor-cross bunch together with 1inch of wire each connecting them in the three places connecting the eight bunchs of four-Capacitor-cross's together per mono Aleph J-X, so there will be no more than 4inchs or less than 1inch at the most of wire used to link all 16 Capacitors per bunch together to the point of either the inductor or the resistor in the CLCRCRC Power Supply, the buchs of the Capacitors will be 3D though, it might look strange eight three dimentional clumps of 16 Capacitors per mono AJ-X, mabe it will look cool to though, or maybe it will be like a work of art, or maybe I should forget about this, but it feels so right ?
Last edited:
choosing electrlytic capacitors
I have found an electrlytic capacitor with good data sheet specs compared to the Panasonic 25v 6800microFerrids TSHA series $1.52 each
http://www.panasonic.com/industrial/components/pdf/pic_ts-ha_dne.pdf
it is the Nichlicon 25v 6800microFerrids PW series $1.17 each
http://nichicon-us.com/english/products/pdf/e-pw.pdf
Look over the two and let me know if I should order the Nichlicon because the delivery date is the end of september at Digikey if I order now, do you think it is worth waiting for based on the two data sheets links above ? Or have I made a mistake by thinking that the Nichicon specs are so much better than the Panasonic, three times longer life and significantly lower ripple numbers.
Thank You for your continueing help, I know this is allot of education I am getting here by being able to talk to someone about this project.
I have found an electrlytic capacitor with good data sheet specs compared to the Panasonic 25v 6800microFerrids TSHA series $1.52 each
http://www.panasonic.com/industrial/components/pdf/pic_ts-ha_dne.pdf
it is the Nichlicon 25v 6800microFerrids PW series $1.17 each
http://nichicon-us.com/english/products/pdf/e-pw.pdf
Look over the two and let me know if I should order the Nichlicon because the delivery date is the end of september at Digikey if I order now, do you think it is worth waiting for based on the two data sheets links above ? Or have I made a mistake by thinking that the Nichicon specs are so much better than the Panasonic, three times longer life and significantly lower ripple numbers.
Thank You for your continueing help, I know this is allot of education I am getting here by being able to talk to someone about this project.
scaiff001,
I am constructing a multiple capacitor power supply for my current amp. I found some nice copper bars which I will run down the middle of the caps and solder each to the bar. It will hold everything together and has incredible current carrying capacity. You can just drill a small hole for each electrode and solder on the top. I plan to mount the R resistors to this bar also, and run the plus and minus poles down each side. I am using 14 8200 uf caps from Apex Jr.
The above idea is just in case you do not have a nice fabricated pcb to do the same job.
Tad
I am constructing a multiple capacitor power supply for my current amp. I found some nice copper bars which I will run down the middle of the caps and solder each to the bar. It will hold everything together and has incredible current carrying capacity. You can just drill a small hole for each electrode and solder on the top. I plan to mount the R resistors to this bar also, and run the plus and minus poles down each side. I am using 14 8200 uf caps from Apex Jr.
The above idea is just in case you do not have a nice fabricated pcb to do the same job.
Tad
Don't wait, you'll drive yourself nuts second guessing yourself. Either one will handle the ripple current with a nice margin, which you have verified by using PSU designer, right?
in case you missed it earlier USE 35V CAPS - YOU WILL EXCEED 25V ON SOME OF THEM. Here's the post http://www.diyaudio.com/forums/pass-labs/145387-aleph-j-x-amp-project-38.html#post2239050
in case you missed it earlier USE 35V CAPS - YOU WILL EXCEED 25V ON SOME OF THEM. Here's the post http://www.diyaudio.com/forums/pass-labs/145387-aleph-j-x-amp-project-38.html#post2239050
Scaiff001
Both caps are a good choice. If you are running 25V rails you will want to make sure that the 25V on the cap is a WV (working voltage) or multiply the voltage of your rails by 1.25 to get the minimum voltage of the caps. Always have at least a 25% margin above the voltage of your rails. Yes, WV should be safe at the same voltage but that rating of 3000hrs will come into play a lot faster the closer the cap voltage rating is to the actual voltage the cap passes.
Uriah
Both caps are a good choice. If you are running 25V rails you will want to make sure that the 25V on the cap is a WV (working voltage) or multiply the voltage of your rails by 1.25 to get the minimum voltage of the caps. Always have at least a 25% margin above the voltage of your rails. Yes, WV should be safe at the same voltage but that rating of 3000hrs will come into play a lot faster the closer the cap voltage rating is to the actual voltage the cap passes.
Uriah
The filter input caps will see 27V with the transformer voltages chosen, before allowing for transformer regulation, voltage tolerance or high mains voltage. Good design practice requires 35V caps, even though some people would run 25V caps here. As Uriah points out, for reliability go with the 35V caps, at least for the first CLC section. At 200 pieces, there's only a $.20/cap difference
Last edited:
http://www.diyaudio.com/forums/apex-jr/165820-fs-panasonic-18-000uf-50v-snap.html#post2250086
I think there is no better purchase 🙂
I think there is no better purchase 🙂
there is a problem with the mathamatics we use not the soundness of the design
the amp should be built, the problem is not in the design itself but in the constant errors produced by the current mathamatics of mankind, I am working on a General Counting Theory that should solve all of the current math problems, repeating the numbers 0-9 over and over again is nieve behavior of people because we have ten fingers and toes, this dose not produce an acceptable counting theory that currently destroys other mathamatical axioms including those used in electoacoustics along with other areas of physics.
IMHO, the issue is less whether we call that first finger 0, 1 or z than it is that the end result in audio is subjective. Even though there are a number of theories about what distortions are audible, in the end it is up to each of us to decide if any improvement is audible and worth the expense of achieving it.
A wealthy friend is perfectly happy with his mid priced receiver driving a pair of 30" high "towers" with 2 x 4" woofers and a 12" 150W PMPO sub in a 1500 sq ft room. It is almost unlistenable to me at anything more than background levels. He acknowledges that my system sounds a bit cleaner, but to him that front projector and 100" screen were the place to put his money. Not a choice I think either of us would make.
The point is that it is up to you to try things and decide if you like the sonic effect. To me your proposed PSU is way past overkill, but you wouldn't be the first to build it that way. I've tried to help you learn a bit about specifying parts for reliability, leaving the sonic merits up to you.
I don't think you'll be disappointed with any Pass design. If not for the trouble my air conditioning is having keeping up with the heat load this summer, I'd be listening with to my A75. I've got Aleph and AX boards and parts, but haven't had enough of an issue with the A75 to get to building them up. Looking forward to Fall and cool enough weather to go back to class A.
Happy building and listening.
A wealthy friend is perfectly happy with his mid priced receiver driving a pair of 30" high "towers" with 2 x 4" woofers and a 12" 150W PMPO sub in a 1500 sq ft room. It is almost unlistenable to me at anything more than background levels. He acknowledges that my system sounds a bit cleaner, but to him that front projector and 100" screen were the place to put his money. Not a choice I think either of us would make.
The point is that it is up to you to try things and decide if you like the sonic effect. To me your proposed PSU is way past overkill, but you wouldn't be the first to build it that way. I've tried to help you learn a bit about specifying parts for reliability, leaving the sonic merits up to you.
I don't think you'll be disappointed with any Pass design. If not for the trouble my air conditioning is having keeping up with the heat load this summer, I'd be listening with to my A75. I've got Aleph and AX boards and parts, but haven't had enough of an issue with the A75 to get to building them up. Looking forward to Fall and cool enough weather to go back to class A.
Happy building and listening.
I have found these two Electrolytic Capacitors with the first having a long life for people who listen to music allot, it is 250,000 hours Ripple number of 9.2
Vishay 33KuF part number 56333E3 or 16333E3 the data sheet follows this
http://www.farnell.com/datasheets/415595.pdf
The second one having a life expectancy of 20,000 hours and a Ripple number of 26.4 BHC Aerovox 33KuF part number ALS3-334RH025 the data sheet follows this
http://www.farnell.com/datasheets/24905.pdf
The price of these capacitors will dictate the type of P.S. I will build, it will have to be simple with as few capacitors as posible, maybe CLC ?
Vishay 33KuF part number 56333E3 or 16333E3 the data sheet follows this
http://www.farnell.com/datasheets/415595.pdf
The second one having a life expectancy of 20,000 hours and a Ripple number of 26.4 BHC Aerovox 33KuF part number ALS3-334RH025 the data sheet follows this
http://www.farnell.com/datasheets/24905.pdf
The price of these capacitors will dictate the type of P.S. I will build, it will have to be simple with as few capacitors as posible, maybe CLC ?