Bear in mind that an alternate way to diddle the DC offset is with the 392 ohm load resistors for the front end differential. Just set up a pair of resistors and a pot (to replace each 392), not unlike the rig for the current source. That way you can tweak one side if you wish.
There's no law that says the Zener for the current source has to be 9.1V. If you want to run rails less than, say, 10 or 12V consider using a smaller value Zener, perhaps 6V. Be prepared to reduce the value of the resistor under the Source of the current source, however, to reflect the reduced voltage. All you need is something on the order of 3 or 4V at the Gate (Vgs) of the current source to turn it on, so there's still plenty of leeway if you're willing to juggle the values a bit.
You can get the rails down pretty low with very little effort. The topology won't need to change until you get below 5 or 6V rails.
Horowitz & Hill present a nifty current source for those who want a really high performance circuit. There are scads and scads of other current sources out there; the one known locally as the 'HH' isn't really new at all. You can find a number of variations on it if you look around...as well as many, many others. You can make a current sources a life-long obsession if you want. I used to collect current source schematics until I found this really neat one, printed it off, then leafed back through all the others I had and discovered that I'd already 'found' it twice before. <i>Ahem.</i> After that, I figured I'd be better off obsessing about something else.
Grey
There's no law that says the Zener for the current source has to be 9.1V. If you want to run rails less than, say, 10 or 12V consider using a smaller value Zener, perhaps 6V. Be prepared to reduce the value of the resistor under the Source of the current source, however, to reflect the reduced voltage. All you need is something on the order of 3 or 4V at the Gate (Vgs) of the current source to turn it on, so there's still plenty of leeway if you're willing to juggle the values a bit.
You can get the rails down pretty low with very little effort. The topology won't need to change until you get below 5 or 6V rails.
Horowitz & Hill present a nifty current source for those who want a really high performance circuit. There are scads and scads of other current sources out there; the one known locally as the 'HH' isn't really new at all. You can find a number of variations on it if you look around...as well as many, many others. You can make a current sources a life-long obsession if you want. I used to collect current source schematics until I found this really neat one, printed it off, then leafed back through all the others I had and discovered that I'd already 'found' it twice before. <i>Ahem.</i> After that, I figured I'd be better off obsessing about something else.
Grey
Absolutely agreed, Grey. One could go batty, fretting over current sources. Apart from the addition of a negative-rail connected FET to feed the reference diode, and slightly different component choices, the "HH" current source is identical to your original. With very low rail voltages, one could potentially even switch to a bipolar current source arrangement, in order to retain sufficient voltage and a bit of headroom for the proper operation of the diff pair.
Now, while I <i>could</i> go and tweak drain resistances to zero the DC across the load, I think it's preferable to try and find a matched diff pair first. That way, the transistors and resistors are both matched, rather than depending on a mismatch in one to compensate for a mismatch in the other...
Peter:
How many 9610s did you have to go through to find a pair that matched well enough to give you only 3mV of offset?
Now, while I <i>could</i> go and tweak drain resistances to zero the DC across the load, I think it's preferable to try and find a matched diff pair first. That way, the transistors and resistors are both matched, rather than depending on a mismatch in one to compensate for a mismatch in the other...
Peter:
How many 9610s did you have to go through to find a pair that matched well enough to give you only 3mV of offset?
As I said, I had them matched out of circuit first. I think I needed no more than 4 close matched pairs to get one perfectly matched pair. Altogether I bought 25 mosfets and I have enough for future projects.
one known locally as the 'HH' isn't really new at all.
It was new when I designed it...... There is more to designing a current source than the basic topology choice. Output impedance
,low capacitance, device choices, and PSRR for the voltage reference were the parameters I lookat. There was nothing at all obsessive about it and the only inspiration to do it came from the very poor PSRR for the voltage refence of the origional Aleph X current source design which has been discussed by others as well on the forum. I actually have some other ideas on the subject but would not want to be called obsessive or confuse anybody further. Welcome back Grey and thanks for your usual insightful analysis.
It was new when I designed it...... There is more to designing a current source than the basic topology choice. Output impedance
,low capacitance, device choices, and PSRR for the voltage reference were the parameters I lookat. There was nothing at all obsessive about it and the only inspiration to do it came from the very poor PSRR for the voltage refence of the origional Aleph X current source design which has been discussed by others as well on the forum. I actually have some other ideas on the subject but would not want to be called obsessive or confuse anybody further. Welcome back Grey and thanks for your usual insightful analysis.
Current Source Blues
The HH cpleaseurrent source is clearly an improvement over the original for anyone that has cared to build it. The key here is to try it before you get on your soapbox.......
There are a circuits that are a lot more complex that offer better performance on paper that sound inferior to the orignal. The differences lie in the details and correct implementation of the circuit. A badly designed current source can ruin an otherwise good circuit. This is probably why a certain segment of our community hate current sources.
Fred, please feel free to discuss your other ideas as there are some of us that would like to explore ideas and improvements of designs ....... or maybe we should sit back and relax, after all ignorance is bliss.
Jam
The HH cpleaseurrent source is clearly an improvement over the original for anyone that has cared to build it. The key here is to try it before you get on your soapbox.......
There are a circuits that are a lot more complex that offer better performance on paper that sound inferior to the orignal. The differences lie in the details and correct implementation of the circuit. A badly designed current source can ruin an otherwise good circuit. This is probably why a certain segment of our community hate current sources.
Fred, please feel free to discuss your other ideas as there are some of us that would like to explore ideas and improvements of designs ....... or maybe we should sit back and relax, after all ignorance is bliss.
Jam
back to square one - an elementary printing question
Grey posted a schematics on the 1st post on this thread. How do you guys print out the .bmp file so that it'll fit on one 8 1/2 X11 paper? So far all the print out that I got are either too small or too big (that requires 4 pages) and the text is unreadable. I have used MS paint and IrfanView.
I want to have a hardcopy to compare the differences between the schematics that were posted by Grey and Chad (schematics with a couple of options - in another thread).
Grey posted a schematics on the 1st post on this thread. How do you guys print out the .bmp file so that it'll fit on one 8 1/2 X11 paper? So far all the print out that I got are either too small or too big (that requires 4 pages) and the text is unreadable. I have used MS paint and IrfanView.
I want to have a hardcopy to compare the differences between the schematics that were posted by Grey and Chad (schematics with a couple of options - in another thread).
Fcel,
The way I did mine was to open it in MS-paint, select all, edit-copy, open a blank doc in word Landscape layout, paste in doc, use handle to drag it as big enough to fill the A4 size, edit-select all, make it bold, and print doc.
Though it is very fine and faint, but just able to read all values and parts number.
Chris
The way I did mine was to open it in MS-paint, select all, edit-copy, open a blank doc in word Landscape layout, paste in doc, use handle to drag it as big enough to fill the A4 size, edit-select all, make it bold, and print doc.
Though it is very fine and faint, but just able to read all values and parts number.
Chris
Since Paint doesn't seem to let you zoom to any size less than 100%, I opened it with Imaging for Windows. That came with Windows 98 that I'm running. It was under Programs / accessories. That's a more flexible program and will let me zoom to any size I want. Other image handling programs, maybe photo editing programs, might work for this too.
<center><i><font color="#800080#000000" face="Fantasy">How do you guys print out the .bmp file
so that it'll fit on one 8 1/2 X11 paper?</font></i></center><hr width="95%" align=center>
I use <a href="http://www.irfanview.com/">irfanview</a> it will fit to the page size used by your printer, with adjustable borders.
It also hase a very good resizing algorythm, opens most file types, and its free.
Hope this helps
Regards
James
so that it'll fit on one 8 1/2 X11 paper?</font></i></center><hr width="95%" align=center>
I use <a href="http://www.irfanview.com/">irfanview</a> it will fit to the page size used by your printer, with adjustable borders.
It also hase a very good resizing algorythm, opens most file types, and its free.
Hope this helps
Regards
James
Re: back to square one - an elementary printing question
I used Infranview and FindPrint to print the schematics
and they came out fine. Unfortunately I cannot check
the exact settings used since my main computer blew
up a few days ago. 🙁
Dennis
I used Infranview and FindPrint to print the schematics
and they came out fine. Unfortunately I cannot check
the exact settings used since my main computer blew
up a few days ago. 🙁
Dennis
InfranView print out is not as good as Imaging for Windows. This is how I did it (finally):
Image->Resize->1600X1200->Image->Sharpen (a couple of times) and then for printing, select "best fit to page".
Image->Resize->1600X1200->Image->Sharpen (a couple of times) and then for printing, select "best fit to page".
Jens,
I tried opening the file but I am being told it is a MS Picture It! file, which I don't have. It says zip but both my utilities can't decompress it. Can you try posting it again?
I tried opening the file but I am being told it is a MS Picture It! file, which I don't have. It says zip but both my utilities can't decompress it. Can you try posting it again?
I´m really curious on the sound from a AX.
Who to contact if I want to buy a couple of boards?
I tried to contact HifiZen with a PM, is that the right fellow?
I tried the Aleph5 a couple of years ago but in the end clearly prefered my class A Zapsolute. However, there was something very nice ( as if you guys did´nt know that 🙂 about the A5, especially the warm and relaxed midrange. However it lacked the deep controled bass of my amps, didn´t put up the same clear soundstage, and didn´t have the full resolution in the highest octave. Also my 78W monos had better drive/dynamics with complex blues & rock.
Maybe the AlephX or XA series is what I have been looking for for a couple of years. The Gammut D200 and ARC100mk2 didn´t really make it either, even if they both were extremly nice as was the Aleph5.
What I really would like to see from Mr Pass is a class A design, fully symmetric and free from feedback 🙂
Thanks!
/Peter
Who to contact if I want to buy a couple of boards?
I tried to contact HifiZen with a PM, is that the right fellow?
I tried the Aleph5 a couple of years ago but in the end clearly prefered my class A Zapsolute. However, there was something very nice ( as if you guys did´nt know that 🙂 about the A5, especially the warm and relaxed midrange. However it lacked the deep controled bass of my amps, didn´t put up the same clear soundstage, and didn´t have the full resolution in the highest octave. Also my 78W monos had better drive/dynamics with complex blues & rock.
Maybe the AlephX or XA series is what I have been looking for for a couple of years. The Gammut D200 and ARC100mk2 didn´t really make it either, even if they both were extremly nice as was the Aleph5.
What I really would like to see from Mr Pass is a class A design, fully symmetric and free from feedback 🙂
Thanks!
/Peter
"I'm using a circuit to eliminate DC component from AC line. It's taken from ML 23.5. Indeed, when I was using it in A75, the tranformers were extremally quiet without a smallest hum. The power is switched through a triac by a togle switch and remotedly by wire using 12V relay. I used two CL-70 thermistors connected in series for inrush protection. It may be overkill, but I'm tired of changing bridges.
Peter, do you have a schematic on this circuit or can you elaborete a bit more on that ? I have seen that Nelson basically suggests Variacs in the primary to reduce DC. The circuit of LC Audio using diodes / caps seem not to be the best solution as well (Jens' comments earlier), so I wonder how your setup looks like.
Thanks & Best Regards
Peter, do you have a schematic on this circuit or can you elaborete a bit more on that ? I have seen that Nelson basically suggests Variacs in the primary to reduce DC. The circuit of LC Audio using diodes / caps seem not to be the best solution as well (Jens' comments earlier), so I wonder how your setup looks like.
Thanks & Best Regards