Asen said:
Depends on how you do the PS. If you do it "GC-style" it will sound like a gainclone, but if you build a more normal PSU, it will sound more like a normal amp (more details in the lower region, and more controlled in the upper frequencies, basically more controlled all together. What will sound best in a particular setup depends a lot on the speakers and listening habits. I have noticed that it pays to use large transformers, especially with the non-filtered (GC) style psu. Up to 400VA is an improvement. Also it has a balanced input option, and that makes it sound very nice with a balanced pre, or makes it possible to bridgde 2 amps without any extra gain stages.
Re: CCS BOZ
I'm currently working on the best implementation of a BOZ version for SEWA, stay tuned
Tyimo said:Hi Mads!
Sorry, you are right, your WEB page is realy working.
Something different theme:
Could you suggest me a nice CCS for the BOZ? I would like to try the BOZ with it.
Happy weekend!
Tyimo
I'm currently working on the best implementation of a BOZ version for SEWA, stay tuned
Banned
Joined 2002
Banned
Joined 2002
X2r1 vs optimal SEWA preamp
I have put the X2 Rev.1 design on the shelf for the time being, and are pursuing a BOZ approach to preamplification. The reason being both Tyimo's listening tests, and my own initial comparisons.
I suspect the high output impedance from the X2r1 to be suboptimal for SEWA. I have used the Aleph L Rev.1.2 as a template for my new preamp, which I have called BOSEWA, in keeping with the Zen nomenclature. It will be sort of a BOZ and Aleph L 1.2 hybrid with about 100R output impedance and 19dB Gain. It has +/-25V output swing, and that makes a 40W into 8R SEWA concept possible. I have ordered some professional proto PCB's for this project, and I have also left options on the board to make it possible to make a Aleph L 1.2 clone.
I have put the X2 Rev.1 design on the shelf for the time being, and are pursuing a BOZ approach to preamplification. The reason being both Tyimo's listening tests, and my own initial comparisons.
I suspect the high output impedance from the X2r1 to be suboptimal for SEWA. I have used the Aleph L Rev.1.2 as a template for my new preamp, which I have called BOSEWA, in keeping with the Zen nomenclature. It will be sort of a BOZ and Aleph L 1.2 hybrid with about 100R output impedance and 19dB Gain. It has +/-25V output swing, and that makes a 40W into 8R SEWA concept possible. I have ordered some professional proto PCB's for this project, and I have also left options on the board to make it possible to make a Aleph L 1.2 clone.
Banned
Joined 2002
I am wondering on your schematic that you sent me for this smd sewa amps REVb. I am wondering if i can substitute C3 witch is 4700uf @ 25V for a 1500uf @50V will this cause a problem ?
Also you should put up some information telling people that Z1 Looks exactly like Q3 ( bc850C ) transistor. It was confusing at first till i saw the picture you have of your board 50% completed.
Will it also hurt to run them @ 40V rails ?
jase
Also you should put up some information telling people that Z1 Looks exactly like Q3 ( bc850C ) transistor. It was confusing at first till i saw the picture you have of your board 50% completed.
Will it also hurt to run them @ 40V rails ?
jase
Banned
Joined 2002
Re: X2r1 vs optimal SEWA preamp
Id be interested in a aleph 1.2 Clone wonder if it would be smd I am loving this smd soldering stuff.
Mad_K said:I have put the X2 Rev.1 design on the shelf for the time being, and are pursuing a BOZ approach to preamplification. The reason being both Tyimo's listening tests, and my own initial comparisons.
I suspect the high output impedance from the X2r1 to be suboptimal for SEWA. I have used the Aleph L Rev.1.2 as a template for my new preamp, which I have called BOSEWA, in keeping with the Zen nomenclature. It will be sort of a BOZ and Aleph L 1.2 hybrid with about 100R output impedance and 19dB Gain. It has +/-25V output swing, and that makes a 40W into 8R SEWA concept possible. I have ordered some professional proto PCB's for this project, and I have also left options on the board to make it possible to make a Aleph L 1.2 clone.
Id be interested in a aleph 1.2 Clone wonder if it would be smd
I am loving this smd soldering stuff.
Using 1500uF C3 will cause Low Frequency rolloff. If it will hurt depends on your the LF performance of your speakers. To calculate rolloff vs C3, use this formula: 1/(C*R*2PI)=-3dB rolloff point , which for 1500uF and 8R speaker is: 1/(1500^10-6*8*2x3,14)= -3dB @ 13,27Hz. The original (mine) has 4,23Hz with 4700uF and 8ohm speakers.
40V rails is pushing it. 40V*2A=80W. Thats 40W per mosfet. If you manage to keep the junctions not much above 100'C (50'C heatsink temp) you'd be OK. You'll need a heatsink with (50'C-25'C ambient)/80W )=25/80'C/W= about 0,3'C/W. I'd use a 0,25'C/W heatsink PER CHANNEL for this.
You will also need to use R1=6K2 to keep the output DC level correct about midpoint between the supplies (20V).
You also need to change R5=4K7 to keep Q3 bias current reasonable enough for R6 to handle it.
Ofcourse you also need to use 40V caps. C5 can be lovered in value without ill effect if you have a low ripple PSU (33KuF-0,47R-33KuF per channel).
A sneak-preview of BOSEWA/AlephLr1.2:
http://home.broadpark.no/~viiaudio/BOSEWAr0.jpg
40V rails is pushing it. 40V*2A=80W. Thats 40W per mosfet. If you manage to keep the junctions not much above 100'C (50'C heatsink temp) you'd be OK. You'll need a heatsink with (50'C-25'C ambient)/80W )=25/80'C/W= about 0,3'C/W. I'd use a 0,25'C/W heatsink PER CHANNEL for this.
You will also need to use R1=6K2 to keep the output DC level correct about midpoint between the supplies (20V).
You also need to change R5=4K7 to keep Q3 bias current reasonable enough for R6 to handle it.
Ofcourse you also need to use 40V caps. C5 can be lovered in value without ill effect if you have a low ripple PSU (33KuF-0,47R-33KuF per channel).
A sneak-preview of BOSEWA/AlephLr1.2:
http://home.broadpark.no/~viiaudio/BOSEWAr0.jpg
How about a transformer for voltage gain? Something like the Passive Voltage Amplifier - PVA @ http://www.electra-print.com/electraamp.html
regards,
PSz.
regards,
PSz.
Hi!
I have just built the SEWA, with 40 V supplies. In fact, I modified my ZEN V3. Used its power supply, removed some parts, turned others upside down, and there it was!
The sound was good, comparing to my ZEN V8. It is very hard to say which one is better. They sound different. SEWA sounds more like solid-state. It is solid, but a bit dry. Imaging and a soundstage are excellent. Bass is good, midrange as well. Highs are a bit shy. Could be better. I am using an IRFP250N.
On the other hand, Zen V8 sounds very relaxed and sweet, but sometimes too "loose".
I am getting an impression that a cascoded follower (mosfet-Jfet) could be exactly what I want. It could merge the advantages of both SEWA and ZEN V8. It would sound like SEWA, but with a bit of triode-like flawor, which I like...
It would be nice if someone could help and come up with such a schematic. I have read some articles about cascode followers, but I am not sure if I could bias them properly. Though it shouldn't be too hard for the more knowledgeable...
By the way, one question for Mr. Pass (if he is reading this):
Why none of the ZEN variation was designed like SEWA (common drain)?
Yes, I understand that common source operation provides voltage gain as well, but then you have to apply feedback, and then....ugh...
Personally I don't care about power, 5-10 watts are enough...
Regards,
Vix
I have just built the SEWA, with 40 V supplies. In fact, I modified my ZEN V3. Used its power supply, removed some parts, turned others upside down, and there it was!
The sound was good, comparing to my ZEN V8. It is very hard to say which one is better. They sound different. SEWA sounds more like solid-state. It is solid, but a bit dry. Imaging and a soundstage are excellent. Bass is good, midrange as well. Highs are a bit shy. Could be better. I am using an IRFP250N.
On the other hand, Zen V8 sounds very relaxed and sweet, but sometimes too "loose".
I am getting an impression that a cascoded follower (mosfet-Jfet) could be exactly what I want. It could merge the advantages of both SEWA and ZEN V8. It would sound like SEWA, but with a bit of triode-like flawor, which I like...
It would be nice if someone could help and come up with such a schematic. I have read some articles about cascode followers, but I am not sure if I could bias them properly. Though it shouldn't be too hard for the more knowledgeable...
By the way, one question for Mr. Pass (if he is reading this):
Why none of the ZEN variation was designed like SEWA (common drain)?
Yes, I understand that common source operation provides voltage gain as well, but then you have to apply feedback, and then....ugh...
Personally I don't care about power, 5-10 watts are enough...
Regards,
Vix
Good info Vix, I'm pleased to hear you like it
Nice to hear about comparisons with The Masters designs
Did you use R1=6K2 (20V on Q1 Source)?
What bias current do you have?
I haven't invested much time with the power J-Fets (as I am too busy with mosfet designs). Besides, Nelson and Grey is doing such a great job I was thinking about cascoding for SEWA, but I'm enjoying it so much as it is right now. From experience, I can imagine that The Sound Of SEWA (which is very little) would clean up a tad more (more "dry-ness"). BTW, I don't find SEWA as I have built it to sound "dry" at all.. I'm not completely shure I would find cascoding more musically pleasing. Also, I don't think cascoding has such a great impact on a follower as it has on a common source design. It has much lower distortion to start with (before feedback anyway). Only one way to find out.. I do have it in the back of my mind (along with a stack of other things to try out).
Anyone else managed to build one yet? I'd love to hear about it, and of course some pictures would say a thousand words...
Nice to hear about comparisons with The Masters designs
Did you use R1=6K2 (20V on Q1 Source)?
What bias current do you have?
I haven't invested much time with the power J-Fets (as I am too busy with mosfet designs). Besides, Nelson and Grey is doing such a great job
I was thinking about cascoding for SEWA, but I'm enjoying it so much as it is right now. From experience, I can imagine that The Sound Of SEWA (which is very little) would clean up a tad more (more "dry-ness"). BTW, I don't find SEWA as I have built it to sound "dry" at all.. I'm not completely shure I would find cascoding more musically pleasing. Also, I don't think cascoding has such a great impact on a follower as it has on a common source design. It has much lower distortion to start with (before feedback anyway). Only one way to find out.. I do have it in the back of my mind (along with a stack of other things to try out). Anyone else managed to build one yet? I'd love to hear about it, and of course some pictures would say a thousand words...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.