Hi Vinnie,
The BOSOZ refers to the NP designed Balanced Line Stage.
Balanced Zen Line Stage
The X refers to Henrik's implementation of NP's supersymmetry feedback to the original BOSOZ circuitry. In a balanced system some feedback from one side goes to the opposite side. Hence the X.
The original design biases the mosfet using a passive current source (a resistor tied to the negative supply). An optional active constant current source can be used to raise the psrr and to give better unbalanced -> balanced conversion. Some say that the passive sounds better.
With this board, one can try both.
Dale
P.S. Do a search for the X-BOSOZ...
The BOSOZ refers to the NP designed Balanced Line Stage.
Balanced Zen Line Stage
The X refers to Henrik's implementation of NP's supersymmetry feedback to the original BOSOZ circuitry. In a balanced system some feedback from one side goes to the opposite side. Hence the X.
The original design biases the mosfet using a passive current source (a resistor tied to the negative supply). An optional active constant current source can be used to raise the psrr and to give better unbalanced -> balanced conversion. Some say that the passive sounds better.
With this board, one can try both.
Dale
P.S. Do a search for the X-BOSOZ...
First cut optimization of Apox-1 resistor values
I have made some changes to Apox-1 rev 3 Excel spreadsheet for computing optimal values of the attenutator.
Nice work Craig!
I got some surprises along the way and used goal-seek extensively to figure out what was going on.
In the end I came up with a setup which might be pretty good. I am quite torn as to how I should work this -- should one have a relatively high fixed series resistor, or should one have this resistor smaller than the next parallell unit that is switched in? Hmmm.
Check it out and let me know what you think
Petter
I have made some changes to Apox-1 rev 3 Excel spreadsheet for computing optimal values of the attenutator.
Nice work Craig!
I got some surprises along the way and used goal-seek extensively to figure out what was going on.
In the end I came up with a setup which might be pretty good. I am quite torn as to how I should work this -- should one have a relatively high fixed series resistor, or should one have this resistor smaller than the next parallell unit that is switched in? Hmmm.
Check it out and let me know what you think
Petter
Re: APOX-3 schematic rollout
Well how much just for the PCB do you figure Craig?
I would be interested in two myself. What do you think of Mono X-SOBOZ boards? Then I would want four boards!
Regards
Anthony
🙂
dipchip said:
Well how much just for the PCB do you figure Craig?
I would be interested in two myself. What do you think of Mono X-SOBOZ boards? Then I would want four boards!
Regards
Anthony
🙂
Anthony,
Would you rather have mono boards or two mono circuits on one board (separate power supplies) ?
What about the power supply board:
one with two independant circuits or two boards?
Dale
Would you rather have mono boards or two mono circuits on one board (separate power supplies) ?
What about the power supply board:
one with two independant circuits or two boards?
Dale
Dale, Craig,
I have followed this with fascination - congratulations for producing such a fine implementation. This stuff is not easy, and as you we all know, pleasing everyone is not possible!
I originally went down this route, and did all the math with 11 relays and about 58dB of attenuation, in carefully measured steps. It was a combination of shunt and series, and would have required a lot of work to complete, with unknown outcomes. Thus I was reluctant to take that step, and cast around for a very good motorized pot, finally settling on the scarce-as-hens-teeth EUWM from Panasonic, which is absolutely exceptional. I then figured that the reason pots have such a profound influence on the sound is because the source impedance is too high (noise issues), and the target impedance too low (current, and noise issues). I reasoned that putting the pot after a buffer and working it into a very high impedance tube grid would be the go - and this led to the GK-1. I also went down the route of cermets from Sfernice, which are also very good but suffer highish wear rates, and eventually determined that this approach gave me the quality required. Your APOX is the very next step.........
Dale, as the digital man, you would be well aware of the programming issues. I fell in with a digital design genius here in Melbourne, Ben Williams, who easily surmounted the odds, and we are about to go with the full remote control for the GK-1. It uses a toggle remote for five way source selection, HT Bypass, Mute and up/down for volume control. It uses the ATMEL 4433 series, with around 450 lines of assembler, and a 4HMz Hi Q crystal. The relays, NAIS Aromat 5V coil four poles, switch ground as well as signal, and are driven via a diode-protected ULN2003A darlington array. We do not use an LCD readout; shame really, but they are quite expensive. I imagine this is all very similar to the APOX control system.
My sincere congratulations on your kit! You have contributed to the art, given something to the DIY community, and started a company and a website. Huge steps, the sky is the limit!! It's a pity we can't all do our projects at the same time, and trade design issues!
Cheers,
Hugh
I have followed this with fascination - congratulations for producing such a fine implementation. This stuff is not easy, and as you we all know, pleasing everyone is not possible!
I originally went down this route, and did all the math with 11 relays and about 58dB of attenuation, in carefully measured steps. It was a combination of shunt and series, and would have required a lot of work to complete, with unknown outcomes. Thus I was reluctant to take that step, and cast around for a very good motorized pot, finally settling on the scarce-as-hens-teeth EUWM from Panasonic, which is absolutely exceptional. I then figured that the reason pots have such a profound influence on the sound is because the source impedance is too high (noise issues), and the target impedance too low (current, and noise issues). I reasoned that putting the pot after a buffer and working it into a very high impedance tube grid would be the go - and this led to the GK-1. I also went down the route of cermets from Sfernice, which are also very good but suffer highish wear rates, and eventually determined that this approach gave me the quality required. Your APOX is the very next step.........
Dale, as the digital man, you would be well aware of the programming issues. I fell in with a digital design genius here in Melbourne, Ben Williams, who easily surmounted the odds, and we are about to go with the full remote control for the GK-1. It uses a toggle remote for five way source selection, HT Bypass, Mute and up/down for volume control. It uses the ATMEL 4433 series, with around 450 lines of assembler, and a 4HMz Hi Q crystal. The relays, NAIS Aromat 5V coil four poles, switch ground as well as signal, and are driven via a diode-protected ULN2003A darlington array. We do not use an LCD readout; shame really, but they are quite expensive. I imagine this is all very similar to the APOX control system.
My sincere congratulations on your kit! You have contributed to the art, given something to the DIY community, and started a company and a website. Huge steps, the sky is the limit!! It's a pity we can't all do our projects at the same time, and trade design issues!
Cheers,
Hugh
My recommendation on the APOX 1 board regarding the resistor selection is to determine a "normalized" set of resistors. IE, you would define a set with 5K minimum impedance, and specify all the resistors. The user would then scale all the resistors to the impedance he needed.
From a quick glance at the excel spreadsheet, it appears that there are a LOT of overlapping settings.
Sheldon
From a quick glance at the excel spreadsheet, it appears that there are a LOT of overlapping settings.
Sheldon
I should add that if you specify a set of "nominal" resistors for the apox1, then you can pick the curve transistons easily and sort of "pre-set" the controller board before sending it out. You'll get a lot fewer questions that way.
Sheldon
Sheldon
harvardian said:
Well I was thinking of two completely discrete Preamps fed from the Left and Right outputs of APOX. This would put each Preamp at each speaker with the amps.
Too Wierd?
Anthony
Coulomb said:
Useless at least.😉 Why bother with preamp in such setup, when amp can provide enough gain. It's unnecessary extra stage, that messes the sound. The whole idea of an active preamp is that it takes away the task of driving the volume cantrol from the source (as well as providing gain if needed)
Sheldon
"A lot of overlapping setting".
This is true, but consider why it has been set up like that and that there are in fact 256 (yes, not 255) discrete settings per range which is more than most people will usually give you 🙂
So back to why it is set up like this -- the reason is that one typically wants to get a low impedance signal into the next stage. A potentiometer is excellent at this as you move up in volume. However, at low volume you get increasingly poorer result.
Back to Apox-1. I believe that in principle the Apox-1 is set up for good sound both at low and high volume settings. You can reduce the number of overlapping settings by increasing the dB gain per transition.
I also subscribe to the pre-selected resistor setup. This probably would reduce to specifying one resistor and have the rest autogenerated -- which could make the code universal.
Petter
"A lot of overlapping setting".
This is true, but consider why it has been set up like that and that there are in fact 256 (yes, not 255) discrete settings per range which is more than most people will usually give you 🙂
So back to why it is set up like this -- the reason is that one typically wants to get a low impedance signal into the next stage. A potentiometer is excellent at this as you move up in volume. However, at low volume you get increasingly poorer result.
Back to Apox-1. I believe that in principle the Apox-1 is set up for good sound both at low and high volume settings. You can reduce the number of overlapping settings by increasing the dB gain per transition.
I also subscribe to the pre-selected resistor setup. This probably would reduce to specifying one resistor and have the rest autogenerated -- which could make the code universal.
Petter
Overlap
Sheldon,
If you mess with the permanent shunt resistor.
(By making it a lower value. you can start to flatten out the
curve, but you also don't start out at zero attenuation.) This might be O.K. if you had a high gain pre-amp.
I also agree that the best way for my firmware to work is to pick some base set of resistors at a low input impedance. Then people who want a higher input impedance would use some mulitplication factor to change all of the resistors.
In fact, if I do this. I can just use the base values in my firmware to calculate the nearest db step.
Coulomb,
I'm not sure on pricing yet, I'll have to see what the final dimensions of the board are. But we will certainly offer a competitive price for the raw board.
I almost wish we sold only raw boards, as I am dreading the
upcoming day of kitting that Dale and I will soon endure.
I now appreciate the shipping/purchasing department at my work more than ever.
-Craig Beiferman
Sheldon,
If you mess with the permanent shunt resistor.
(By making it a lower value. you can start to flatten out the
curve, but you also don't start out at zero attenuation.) This might be O.K. if you had a high gain pre-amp.
I also agree that the best way for my firmware to work is to pick some base set of resistors at a low input impedance. Then people who want a higher input impedance would use some mulitplication factor to change all of the resistors.
In fact, if I do this. I can just use the base values in my firmware to calculate the nearest db step.
Coulomb,
I'm not sure on pricing yet, I'll have to see what the final dimensions of the board are. But we will certainly offer a competitive price for the raw board.
I almost wish we sold only raw boards, as I am dreading the
upcoming day of kitting that Dale and I will soon endure.
I now appreciate the shipping/purchasing department at my work more than ever.
-Craig Beiferman
Hello Craig, whatever configuration you settle on for the X-SOBOZ is fine with me. As far as purchasing nightmares, try runniing a purchasing department for a Pharmaceutical research company! 🙂
Thanks
Anthony
Hi Hugh,
I was hoping that you see the thread and post your opinion. Actually, I felt a little guilty since your project inspired me to start this one. I felt as though we were going after different goals, so I absolved myself of guilt 🙂
Funny how ours designs are so close... We decided to go fully balanced (if desired) and include a LCD (only USD$17).
We are using the PIC micros and writing in C instead of assembler. The speed of development makes up for the loss in optimization. We also use the ULN2803 and switch the grounds.
We also decided to make the system very modular by separating the boards into:
Front panel
Input select
Volume
This allows us to offer many options and future enhancements. We communicate using I2C. Thus, every board has a micro on it.
We now have three volume designs:
Series, shunt, and digital pot. If a gain stage is desired, one can put it before or after the volume board.
The digital pot version (probably PGA2310) may have an oprional gain stage since it has a opamp stage already.
I now understand your complaints about sourcing parts. It is a difficult task and you have to buy in bulk to lower costs so that you at least break even 😱
I ended up buying a software package to help with the purchasing www.trilogydesign.com Not much more than a front end to a JET database, but makes it easier to keep track of parts. We have tried to have many shared parts on each board. The software makes it easier to decide how many parts to order and keep track of stock.
Best Regards,
Dale
I was hoping that you see the thread and post your opinion. Actually, I felt a little guilty since your project inspired me to start this one. I felt as though we were going after different goals, so I absolved myself of guilt 🙂
Funny how ours designs are so close... We decided to go fully balanced (if desired) and include a LCD (only USD$17).
We are using the PIC micros and writing in C instead of assembler. The speed of development makes up for the loss in optimization. We also use the ULN2803 and switch the grounds.
We also decided to make the system very modular by separating the boards into:
Front panel
Input select
Volume
This allows us to offer many options and future enhancements. We communicate using I2C. Thus, every board has a micro on it.
We now have three volume designs:
Series, shunt, and digital pot. If a gain stage is desired, one can put it before or after the volume board.
The digital pot version (probably PGA2310) may have an oprional gain stage since it has a opamp stage already.
I now understand your complaints about sourcing parts. It is a difficult task and you have to buy in bulk to lower costs so that you at least break even 😱
I ended up buying a software package to help with the purchasing www.trilogydesign.com Not much more than a front end to a JET database, but makes it easier to keep track of parts. We have tried to have many shared parts on each board. The software makes it easier to decide how many parts to order and keep track of stock.
Best Regards,
Dale
Coulomb said:
If you want a _real_ nightmare try being a researcher at MIT.
It makes you want to double your prozac intake. 😉
New Ideas Please!
I was wondering what type of board would be the best addition to the APOX series.
I'd like to design something that is not currently available, or at least enhance currently available products with some type of microcontroller enhanced versions.
I was thinking that it would be neat to design a pre-amp whose design features were settable via the front panel interface.
That way the user could simply switch a front panel setting to hear the difference instantly.
Does anyone on this forum have a need for any particular board?
I'd like to get a bunch of ideas, and figure out the next step.
Also anyone out there who would like design their own I2C controlled boards. I am not opposed to adding the control of your boards to the APOX-IR1 software. I am also not opposed to selling your boards on our web site, and giving you back a commision.
Thanks for the recent orders. But we have quite a few more sales to make before we break even.
Thanks,
Craig Beiferman
I was wondering what type of board would be the best addition to the APOX series.
I'd like to design something that is not currently available, or at least enhance currently available products with some type of microcontroller enhanced versions.
I was thinking that it would be neat to design a pre-amp whose design features were settable via the front panel interface.
That way the user could simply switch a front panel setting to hear the difference instantly.
Does anyone on this forum have a need for any particular board?
I'd like to get a bunch of ideas, and figure out the next step.
Also anyone out there who would like design their own I2C controlled boards. I am not opposed to adding the control of your boards to the APOX-IR1 software. I am also not opposed to selling your boards on our web site, and giving you back a commision.
Thanks for the recent orders. But we have quite a few more sales to make before we break even.
Thanks,
Craig Beiferman
Re: New Ideas Please!
If this is possible, there would be definitely a demand for a nice control board for CD-Pro 2. The board that comes with a transport is basic, with only few controls and the dimentions are not really convenient. Dispaly also isn't that great.
Something like ML for instance.
dipchip said:
If this is possible, there would be definitely a demand for a nice control board for CD-Pro 2. The board that comes with a transport is basic, with only few controls and the dimentions are not really convenient. Dispaly also isn't that great.
Something like ML for instance.
Attachments
Re: Re: New Ideas Please!
Jeeze, Peter, not asking for much are you...
😉 But Dale and Craig did ask for ideas.
Peter Daniel said:
Jeeze, Peter, not asking for much are you...
😉 But Dale and Craig did ask for ideas.
Just taking my chances😉
The reason I asked is that with a large order of CD-Pro transports, they treat a customer as industrial end user and don't supply display PCBs. They also send it in batches of 12 units.
That is also a reason for delay with cd pro group purchase.
The reason I asked is that with a large order of CD-Pro transports, they treat a customer as industrial end user and don't supply display PCBs. They also send it in batches of 12 units.
That is also a reason for delay with cd pro group purchase.
For Craig and Peter
Craig,
Are you done with the fans? How was the food?
I am working on the article... I'd rather be working on the APOX.
I sent you some new photos of the USB-CAN board. They look much better.
By the way, why are you concerned about breaking even. I bought all of the parts
Anyways, I'd like people to buy it when they are ready and hopefully there will some honest reviews of it soon.
The cost is high for people to easily buy in on speculation alone.
We don't want to be Bose and hope people will buy something just because we tell them it is good.
The audio community (especially diy) will support a good product. If it's bad, they won't.
Peter, how many confirmed orders for the CD-Pro kit?
Did you ever give me an idea of a preferred size and layout?
How much would you/others be willing to spend?
Dale
Craig,
Are you done with the fans? How was the food?
I am working on the article... I'd rather be working on the APOX.
I sent you some new photos of the USB-CAN board. They look much better.
By the way, why are you concerned about breaking even. I bought all of the parts
Anyways, I'd like people to buy it when they are ready and hopefully there will some honest reviews of it soon.
The cost is high for people to easily buy in on speculation alone.
We don't want to be Bose and hope people will buy something just because we tell them it is good.
The audio community (especially diy) will support a good product. If it's bad, they won't.
Peter, how many confirmed orders for the CD-Pro kit?
Did you ever give me an idea of a preferred size and layout?
How much would you/others be willing to spend?
Dale
Craig,
Type of board:
I have a need to program TI PCM/DSD 1792 chips.
Since most digital parts these days seem to require digital programming using some form of 2 wire interface, it would be really nice to have a universal design that one could add to circuits, either as a plugin deal (in chich case it should be field programmable) or better still a basic circuit with configurable firmware (with a couple of examples where one could be PCM1792 🙂) that one could either license or have as a carrier for people who want to sell other types of boards through your website (with royalties of course).
Dale,
If you want to increase sales, try a review in www.audioxpress.com, or write the article yourselves. I know they are always in need of material -- and they even pay for articles 🙂. To get maximum coverage, fan it out to a 3 part series (Controller board + input switches + Apox 1, Apox 2, Apox 3) and you have 3 months of free advertising towards a DIY crowd.
You also mention that the volume control can be before or after the gain stage. Shame on you -- it can just as well be between gain stages!!! One of the problems with designing simple and good gain stages is the level translation which forces even NP to use AC coupled circuits. With Apox, you can live comfortably at an elevated DC level, and use another gain stage (with say .5-0dB gain) to get you back to zero DC level. No feedback, just lots of degeneration (check out my posts here http://www.diyaudio.com/forums/showthread.php?threadid=3625&perpage=15&pagenumber=5 and http://www.diyaudio.com/forums/showthread.php?threadid=3626) -- real simple. I have built a variant of this using a dual matched JFET on the input and single JFET's on the output -- the only problem is voltage rails to get low gain (apparrently, most people want 0dB gain on their preamps these days)
Petter
Type of board:
I have a need to program TI PCM/DSD 1792 chips.
Since most digital parts these days seem to require digital programming using some form of 2 wire interface, it would be really nice to have a universal design that one could add to circuits, either as a plugin deal (in chich case it should be field programmable) or better still a basic circuit with configurable firmware (with a couple of examples where one could be PCM1792 🙂) that one could either license or have as a carrier for people who want to sell other types of boards through your website (with royalties of course).
Dale,
If you want to increase sales, try a review in www.audioxpress.com, or write the article yourselves. I know they are always in need of material -- and they even pay for articles 🙂. To get maximum coverage, fan it out to a 3 part series (Controller board + input switches + Apox 1, Apox 2, Apox 3) and you have 3 months of free advertising towards a DIY crowd.
You also mention that the volume control can be before or after the gain stage. Shame on you -- it can just as well be between gain stages!!! One of the problems with designing simple and good gain stages is the level translation which forces even NP to use AC coupled circuits. With Apox, you can live comfortably at an elevated DC level, and use another gain stage (with say .5-0dB gain) to get you back to zero DC level. No feedback, just lots of degeneration (check out my posts here http://www.diyaudio.com/forums/showthread.php?threadid=3625&perpage=15&pagenumber=5 and http://www.diyaudio.com/forums/showthread.php?threadid=3626) -- real simple. I have built a variant of this using a dual matched JFET on the input and single JFET's on the output -- the only problem is voltage rails to get low gain (apparrently, most people want 0dB gain on their preamps these days)
Petter
- Status
- Not open for further replies.