Apox 2 Resistors
Harvardian,
What type & brand of resistors will be used in the resistor-relay attenuation network in the Apox 2? Sorry if this has been asked before. I could find no mention of it.
-UncleJessie
Harvardian,
What type & brand of resistors will be used in the resistor-relay attenuation network in the Apox 2? Sorry if this has been asked before. I could find no mention of it.
-UncleJessie
Hi UncleJessie,
First, who's uncle are you? 🙂
Second, Since this is a kit, the choice is of resistors is largely left up to the builder. We will probably "stock" some decent quality sets of resistors for various impedences.
These will probably be Holco H8/H4. The board also has spacing for the caddock resistors. The Vishay S102 can be used, but the leads will have to be slightly formed.
Resistors seem to be one of those audiophile components in which everyone has a favorite. Also, since there are quite a few different values needed, it would be very hard for us financially to stock lots of flavors.
Any suggestions?
Best Regards,
Dale
P.S. I ordered the APOX micro board on Friday. Hopefully, Craig will be ready to order the input and volume boards early this week. I am in a software frenzy.
First, who's uncle are you? 🙂
Second, Since this is a kit, the choice is of resistors is largely left up to the builder. We will probably "stock" some decent quality sets of resistors for various impedences.
These will probably be Holco H8/H4. The board also has spacing for the caddock resistors. The Vishay S102 can be used, but the leads will have to be slightly formed.
Resistors seem to be one of those audiophile components in which everyone has a favorite. Also, since there are quite a few different values needed, it would be very hard for us financially to stock lots of flavors.
Any suggestions?
Best Regards,
Dale
P.S. I ordered the APOX micro board on Friday. Hopefully, Craig will be ready to order the input and volume boards early this week. I am in a software frenzy.
harvardian said:
Hello dale, when you say ordered the APOX boards, do you mean the working Prototype board for APOX 2?
Regards
Anthony
PS. Thanks for shipping the Aelph Boards yesterday.
Hi Anthony,
Sorry, I should have been more clear.
The APOX system is now made up of three different boards:
1) Front panel board (I call it the micro, but all of the boards have micros).
2) Input select board (4 inputs with bypass) Can use two in system.
3) Volume board (Here is where APOX 1 differs from APOX 2)
APOX-2 Each board supports one channel of a Differential system or two channels of a single ended system. Note, if you want even one differential input, you will need two boards. The APOX-2 is a 24 tap series ladder. Constant impedence - 24 volume steps.
APOX-1. This is the P1.7 version where 8 resistors are swapped between series and shunt duty. One board will handle a differential system. Not constant impedence and 256 volume steps (not truly log, so all steps may not be usable)
I ordered three prototype of the FPB. The previous version works fine, but many changes were made.
1) Use I2C to communicate to the other boards
2) Two encoders w/switches for front panel interface (can have mute, bypass, and power on separate switches if desired)
3) Use of "standard" LCD displays (8 bit interface), with LED backlight and contrast control. Any 116mmX37mm display can be directly mounted on the board. Other displays can be hard wired if desired. Noritake has a VFD with the 116mmx37 footprint if anyone wants that instead of a Blue on Black LCD ($58 vs $20)
Hopefully, I will not find any problems with the boards. They are not very complicated and most of the changes were to peripherals and not the main functionality. I am talking to the LCD with the same interface (on a proto board).
Craig is currently finishing the layout on the input board and the APOX-2 version of the volume board. Then he will probably do the APOX-1 version.
I have asked Craig to do the mechanical layout of our prototype unit. This will not be up to Peter's standards, but will give people an idea how a finished design may look. There are several people in the Boston area that we can use as test subjects...
We will probably be offering an APOX-3 which will use the PGA2310 or other digital pot.
Hope that answers some of the questions,
Best Regards,
Dale
Sorry, I should have been more clear.
The APOX system is now made up of three different boards:
1) Front panel board (I call it the micro, but all of the boards have micros).
2) Input select board (4 inputs with bypass) Can use two in system.
3) Volume board (Here is where APOX 1 differs from APOX 2)
APOX-2 Each board supports one channel of a Differential system or two channels of a single ended system. Note, if you want even one differential input, you will need two boards. The APOX-2 is a 24 tap series ladder. Constant impedence - 24 volume steps.
APOX-1. This is the P1.7 version where 8 resistors are swapped between series and shunt duty. One board will handle a differential system. Not constant impedence and 256 volume steps (not truly log, so all steps may not be usable)
I ordered three prototype of the FPB. The previous version works fine, but many changes were made.
1) Use I2C to communicate to the other boards
2) Two encoders w/switches for front panel interface (can have mute, bypass, and power on separate switches if desired)
3) Use of "standard" LCD displays (8 bit interface), with LED backlight and contrast control. Any 116mmX37mm display can be directly mounted on the board. Other displays can be hard wired if desired. Noritake has a VFD with the 116mmx37 footprint if anyone wants that instead of a Blue on Black LCD ($58 vs $20)
Hopefully, I will not find any problems with the boards. They are not very complicated and most of the changes were to peripherals and not the main functionality. I am talking to the LCD with the same interface (on a proto board).
Craig is currently finishing the layout on the input board and the APOX-2 version of the volume board. Then he will probably do the APOX-1 version.
I have asked Craig to do the mechanical layout of our prototype unit. This will not be up to Peter's standards, but will give people an idea how a finished design may look. There are several people in the Boston area that we can use as test subjects...
We will probably be offering an APOX-3 which will use the PGA2310 or other digital pot.
Hope that answers some of the questions,
Best Regards,
Dale
Are your systems going to be backward compatible or upgradeable? ie if I buy an APOX 2 but want to add an APOX 3 Digital pot?
Regards
Anthony
Regards
Anthony
backwards compatability
Coulomb,
Yes, the hardware will be backwards compatible, since each board has a microcontroller on it, which is I2C controlled. So the hardware interface to each new board will always remain the same.
The remote board (front panel board) will be the master controller, and all of the peripheral boards will be the slaves.
However, depending on the features of the new board, the remote control board may need to have a firmware upgrade to properly display all of the features of the new peripheral board.
The firmware upgrade will be accomplished via an RS-232 cable to a bootloader on the remote board.
Also a RS-232 interface will be used to allow individuals to customize their display.
For example, for the input select board you could list
"CD Player"
"Larry's Stereo"
"My TV"
etc..
I hope that makes it clearer,
Craig Beiferman
Coulomb,
Yes, the hardware will be backwards compatible, since each board has a microcontroller on it, which is I2C controlled. So the hardware interface to each new board will always remain the same.
The remote board (front panel board) will be the master controller, and all of the peripheral boards will be the slaves.
However, depending on the features of the new board, the remote control board may need to have a firmware upgrade to properly display all of the features of the new peripheral board.
The firmware upgrade will be accomplished via an RS-232 cable to a bootloader on the remote board.
Also a RS-232 interface will be used to allow individuals to customize their display.
For example, for the input select board you could list
"CD Player"
"Larry's Stereo"
"My TV"
etc..
I hope that makes it clearer,
Craig Beiferman
Re: backwards compatability
Yes, Quite. Thanks Craig. I look forward to you and Dale offering a final release on the APOX 2 System
Anthony
dipchip said:
Yes, Quite. Thanks Craig. I look forward to you and Dale offering a final release on the APOX 2 System
Anthony
I am very pleased with the way this is going, particularly that there will be an Apox 1 which appears to better suit my needs. What i need to see in order to provide constructive feedback is a schematic (however completed it may or may not be).
Petter
Petter
APOX web site updated, more schematics!
Petter,
I have re-arranged the web site (To prevent some confusion I hope), and posted a new APOX-1 schematic for you!
I have almost completed both the APOX-2, and the APOX-IS1 layout. So I hope I can get started with the APOX-1 sometime this week.
Please look over the latest APOX-1 schematic in the new web pages
Thanks,
Craig Beiferman
Petter,
I have re-arranged the web site (To prevent some confusion I hope), and posted a new APOX-1 schematic for you!
I have almost completed both the APOX-2, and the APOX-IS1 layout. So I hope I can get started with the APOX-1 sometime this week.
Please look over the latest APOX-1 schematic in the new web pages
Thanks,
Craig Beiferman
OK Dale, Craig,
Since the posts have slowed a little and I was ignored the last time I made a suggestion, I'll try again now.
Since the posts have slowed a little and I was ignored the last time I made a suggestion, I'll try again now.
roddyama said:
Apox 1 fundamental errors
Well, errors or not, I don't like what i see in the current Apox 1 schematic. This may well be fundamendal design topology disagreements, but please consider checking out my posting on page 9 (note, this drawing shows only the volume section -- input switching is NOT shown if that was confusing before)
What you are doing requires twice the number of relays necessary (yeah, I know if you want balance control, you can use stereo relays to do what you have set up. but still ....).
One of the arguments against shunting between phases was that this did not reduce common mode noise. Well, guess what balanced inputs are for 🙂
An volume control that introduces any unnecessary activity to ground should be avoided in my opinion. This will also allow the volume to be set at a DC level other than ground which is great if you have a simple 2 stage DC coupled setup.
Also, but not quite as important: There is no high/low level setting -- which you have already made some code around.
May I respectfully strongly encourage you to reconsider the Apox 1 volume topology.
Petter
Well, errors or not, I don't like what i see in the current Apox 1 schematic. This may well be fundamendal design topology disagreements, but please consider checking out my posting on page 9 (note, this drawing shows only the volume section -- input switching is NOT shown if that was confusing before)
What you are doing requires twice the number of relays necessary (yeah, I know if you want balance control, you can use stereo relays to do what you have set up. but still ....).
One of the arguments against shunting between phases was that this did not reduce common mode noise. Well, guess what balanced inputs are for 🙂
An volume control that introduces any unnecessary activity to ground should be avoided in my opinion. This will also allow the volume to be set at a DC level other than ground which is great if you have a simple 2 stage DC coupled setup.
Also, but not quite as important: There is no high/low level setting -- which you have already made some code around.
May I respectfully strongly encourage you to reconsider the Apox 1 volume topology.
Petter
APOX-2 Board layout complete
The APOX-2 Board layout is complete! I'll be sending for proto's today. I will try to finish the rest of the APOX-IS1 (Input Select)board layout today.
roddyama,
Yes, we will have the feature you requested!
Petter,
O.K. I think my previous posting of a C program showed that there wasn't that much of a difference in volume control abilities between the method you showed. And the schematic I posted. In fact, with the series resistors changing as you showed, you can get a better range of settings.
I will post a new schematic to see what the general consensus is. And get some further feedback.
Thanks,
Craig Beiferman
The APOX-2 Board layout is complete! I'll be sending for proto's today. I will try to finish the rest of the APOX-IS1 (Input Select)board layout today.
roddyama,
Yes, we will have the feature you requested!
Petter,
O.K. I think my previous posting of a C program showed that there wasn't that much of a difference in volume control abilities between the method you showed. And the schematic I posted. In fact, with the series resistors changing as you showed, you can get a better range of settings.
I will post a new schematic to see what the general consensus is. And get some further feedback.
Thanks,
Craig Beiferman
Hi Rodd,
I apologize for not responding directly to your suggestion. Here is a snippet from my code.
typedef struct {
int1 bGotoLastChannel;
int1 bGotoLastVolume;
int1 bDoStandbyMode;
int1 bVolumeOffsets;
int1 bSplashMode; // if one stays on, else off after 2 seconds
int1 bSix;
int1 bSeven;
int1 bEight;
} APOX_OPTIONS;
// EEPROM data shadowed in RAM
struct {
int8 iNumberInputs;
APOX_OPTIONS options;
int8 iBrightnessLevel;
int8 iContrastLevel;
int8 iLastChannel;
int8 iLastVolume;
int8 iChannelRollover;
int8 iVolumeRollover;
signed int8 iVolumeOffsets[8];
signed int8 iBalanceOffsets; // defined as ???
} apox_ee_data;
As you can see, the iVolumeOffsets[8] is direct support for your feature. The "int1 bVolumeOffsets" is a bool flag to turn the option on/off.
I assume here is what you would expect:
If channel two had an offset of -4, then when switching to that channel, the current volume setting would be the "volume - 4".
Basically all of the channels would able to have a + or - offset. Perhaps, we dont ever need a positive offset and the values would always mean a negative offset?
On a 24 stepped version, the deltas would likely be very small
Best Regards,
Dale
I apologize for not responding directly to your suggestion. Here is a snippet from my code.
typedef struct {
int1 bGotoLastChannel;
int1 bGotoLastVolume;
int1 bDoStandbyMode;
int1 bVolumeOffsets;
int1 bSplashMode; // if one stays on, else off after 2 seconds
int1 bSix;
int1 bSeven;
int1 bEight;
} APOX_OPTIONS;
// EEPROM data shadowed in RAM
struct {
int8 iNumberInputs;
APOX_OPTIONS options;
int8 iBrightnessLevel;
int8 iContrastLevel;
int8 iLastChannel;
int8 iLastVolume;
int8 iChannelRollover;
int8 iVolumeRollover;
signed int8 iVolumeOffsets[8];
signed int8 iBalanceOffsets; // defined as ???
} apox_ee_data;
As you can see, the iVolumeOffsets[8] is direct support for your feature. The "int1 bVolumeOffsets" is a bool flag to turn the option on/off.
I assume here is what you would expect:
If channel two had an offset of -4, then when switching to that channel, the current volume setting would be the "volume - 4".
Basically all of the channels would able to have a + or - offset. Perhaps, we dont ever need a positive offset and the values would always mean a negative offset?
On a 24 stepped version, the deltas would likely be very small
Best Regards,
Dale
APOX-1 Question
Petter,
Would you want to use seperate SPST relays for the left channel.
and another set of relays for the right channel.
Or do you want DPST relays to control both the left and the right at the same time?
I guess if we seperate them(SPST), we can have some sort of balance control. Although the balance would not be linear as you changed volume. (Since the volume settings is not a linear function)
Thanks,
Craig Beiferman
Petter,
Would you want to use seperate SPST relays for the left channel.
and another set of relays for the right channel.
Or do you want DPST relays to control both the left and the right at the same time?
I guess if we seperate them(SPST), we can have some sort of balance control. Although the balance would not be linear as you changed volume. (Since the volume settings is not a linear function)
Thanks,
Craig Beiferman
Craig,
You could make the balance control work linearly (if you used a lookup table or something like that) but I seriously doubt that people need balance.
More importantly:
1* at bottom
I see your point about the volume control, but I don't understand why you are doing the volume setting the way that you are. Guess it is either very smart or not all that smart 🙂 now that I look at it again. Looks like you get a minimum volume of 8K series with 50 or so ohm shorted to ground and a maximum of 50 or so Ohms with zero shorting to ground. That is actually pretty damn good -- you have sort of made all elements "potential pass elements as well as shorting elements". I think we could find a happy medium if you made the various grounds from the relays avaiable so one could short between top and bottom rather than have to have them grounded (All Gnd1 connected, but not grounded unless done so by on purpose by a jumper!) Hey this is not so bad! This would allow the volume control to be available at an elevated voltage which would be extremely convenient in a DC coupled scenario. Adding the possibility for a permanent series resistive element might also be a good idea to set the system up so that the previous stage never sees 50 or so ohms per phase ....
1* (((My personal philosophy on this matter is that a series pass element with shunt swithing is the best setup. To increase the range, a "high" level setup as discussed can be switched in automatically as the volume setting changes in order to increase the energy sent to the next stage. Now, why do I believe such a model is better? Firstly, the signal only passes through one resistor (on low, two in parallel + one relay on half on high), and any switching is done away from the next stage, rather than in series with the next stage. Thus, the series pass elements can be real expensive Vishay, Riken etc. and the shunt elements are less important. But I can live with the above if done as suggested 🙂)))
Petter
You could make the balance control work linearly (if you used a lookup table or something like that) but I seriously doubt that people need balance.
More importantly:
1* at bottom
I see your point about the volume control, but I don't understand why you are doing the volume setting the way that you are. Guess it is either very smart or not all that smart 🙂 now that I look at it again. Looks like you get a minimum volume of 8K series with 50 or so ohm shorted to ground and a maximum of 50 or so Ohms with zero shorting to ground. That is actually pretty damn good -- you have sort of made all elements "potential pass elements as well as shorting elements". I think we could find a happy medium if you made the various grounds from the relays avaiable so one could short between top and bottom rather than have to have them grounded (All Gnd1 connected, but not grounded unless done so by on purpose by a jumper!) Hey this is not so bad! This would allow the volume control to be available at an elevated voltage which would be extremely convenient in a DC coupled scenario. Adding the possibility for a permanent series resistive element might also be a good idea to set the system up so that the previous stage never sees 50 or so ohms per phase ....
1* (((My personal philosophy on this matter is that a series pass element with shunt swithing is the best setup. To increase the range, a "high" level setup as discussed can be switched in automatically as the volume setting changes in order to increase the energy sent to the next stage. Now, why do I believe such a model is better? Firstly, the signal only passes through one resistor (on low, two in parallel + one relay on half on high), and any switching is done away from the next stage, rather than in series with the next stage. Thus, the series pass elements can be real expensive Vishay, Riken etc. and the shunt elements are less important. But I can live with the above if done as suggested 🙂)))
Petter
From Harvardian's post:
First, who's uncle are you?
Bo, Luke, and Daisy's 🙂
Resistors seem to be one of those audiophile components in which everyone has a favorite. Also, since there are quite a few different values needed, it would be very hard for us financially to stock lots of flavors.
Good point, maybe it is best not to even populate the boards with any resistors at all. If you do, just use really cheap ones to demonstrate the operation of your boards. The Vishay's are nice but pretty pricey.
-UncleJessie
First, who's uncle are you?
Bo, Luke, and Daisy's 🙂
Resistors seem to be one of those audiophile components in which everyone has a favorite. Also, since there are quite a few different values needed, it would be very hard for us financially to stock lots of flavors.
Good point, maybe it is best not to even populate the boards with any resistors at all. If you do, just use really cheap ones to demonstrate the operation of your boards. The Vishay's are nice but pretty pricey.
-UncleJessie
Input Select Board Splitting UP?
Hello all,
I completed a layout of the APOX-IS1 (The input select board).
The board was about 12"*3"
Both Dale and I were very dissatisfied with the board layout.
The sheer number of traces made the board a mess.
Even with careful layout, the board just looks bad.
crosstalk, signal length and excessive vias concerned us.
The only way I can think to solve this, is too split up the Input Select Board into two identical boards. One for the 4 left inputs, and an identical board for the 4 right inputs.
These boards, could then either be stacked on top of one another, or laid side by side.
For 8 input users, two left boards would be stacked, and two right boards would be stacked together.
A nice feature of this is that the left and right channels crosstalk would be greatly reduced, since the signals are on physically different boards, and trace length would be cut in half.
Another strange benefit, is that we probably need at least double the number of input select boards, which will probably lower the price for the raw boards.
Mono people would only need 1 board.
The downside is a few extra parts are needed. but I don't think this cost is prohibitive.
Does anyone object to this?
Thanks,
Craig Beiferman
Hello all,
I completed a layout of the APOX-IS1 (The input select board).
The board was about 12"*3"
Both Dale and I were very dissatisfied with the board layout.
The sheer number of traces made the board a mess.
Even with careful layout, the board just looks bad.
crosstalk, signal length and excessive vias concerned us.
The only way I can think to solve this, is too split up the Input Select Board into two identical boards. One for the 4 left inputs, and an identical board for the 4 right inputs.
These boards, could then either be stacked on top of one another, or laid side by side.
For 8 input users, two left boards would be stacked, and two right boards would be stacked together.
A nice feature of this is that the left and right channels crosstalk would be greatly reduced, since the signals are on physically different boards, and trace length would be cut in half.
Another strange benefit, is that we probably need at least double the number of input select boards, which will probably lower the price for the raw boards.
Mono people would only need 1 board.
The downside is a few extra parts are needed. but I don't think this cost is prohibitive.
Does anyone object to this?
Thanks,
Craig Beiferman
- Status
- Not open for further replies.