Ah, you have forgotten your "Change is Inevitable, Dr Spock"!
As your units are setup, it'll be easy to change the way they operate into different loads, multiples, etc via the PIC manipulation/programming (ie, multiple channel vol controls, etc) You've designed in quite a lot of operating flexibility already.
Also, easy to replace the LDR & PIC modules in case of damage/failure - it can happen, unfortunately.
As your units are setup, it'll be easy to change the way they operate into different loads, multiples, etc via the PIC manipulation/programming (ie, multiple channel vol controls, etc) You've designed in quite a lot of operating flexibility already.
Also, easy to replace the LDR & PIC modules in case of damage/failure - it can happen, unfortunately.
For prototype quantities, it's difficult to beat Oshpark.com. You upload the files onto their website, they process and display the boards in real time for your verification, pay by CC and about 2 week delivery to get 3 pcs. Three 2.5" x 3.5" pcbs are $38 with free postage.
For production qtys, try Quick Turn Circuits in Salt Lake City UT.
Both places produce very high quality boards. QTC has excellent customer service. Don't know about Oshpark CS, I've never had the need to speak to a human, everything is automated.
For production qtys, try Quick Turn Circuits in Salt Lake City UT.
Both places produce very high quality boards. QTC has excellent customer service. Don't know about Oshpark CS, I've never had the need to speak to a human, everything is automated.
I've read about this service as well:
Dirt Cheap Dirty Boards Offers Dirt Cheap PCB Fab
Doesn't sound like the best thing for an Audio PCB, but just thought I would throw it out there.
The only problem with Oshpark is you are very limited in your choices:
Only offer 3 pcs increments
Only 0.0625" thick FR4
Only PURPLE solder mask (looks very cool though)
Only gold flash on the pads.
Only offer 3 pcs increments
Only 0.0625" thick FR4
Only PURPLE solder mask (looks very cool though)
Only gold flash on the pads.
Did you know of this:
LDR3x.V2 Passive Preamp Controller | Tortuga Audio
Looks like someone had the same idea. Someone at the DCB1 thread posted the link. Seems expensive though, but really modular build with optional screen ( too simple with two 7-segment Led's, and rediculously expensive ), apple remote option, encoder option.
Now I don't have a clue about what this should cost, but US 259 for the bare volume control seems a bit high...
Any progress in your project??
LDR3x.V2 Passive Preamp Controller | Tortuga Audio
Looks like someone had the same idea. Someone at the DCB1 thread posted the link. Seems expensive though, but really modular build with optional screen ( too simple with two 7-segment Led's, and rediculously expensive ), apple remote option, encoder option.
Now I don't have a clue about what this should cost, but US 259 for the bare volume control seems a bit high...
Any progress in your project??
Yes, I've been aware of their products. This is their second-generation board, I believe; their first generation appeared from the images to have used some variant of the original Silonex-suggested circuit. It's modular, but pretty much a closed system whereas mine is an open system where you can bring your own pots, IR remote, even set up your own display.
For example, my board can work off a linear pot delivering 0-5V to the PIC. Typical pots have two levels, and my design only uses one, you could use the second level to run any panel voltmeter with simple calibration to give you a very large numerical volume level display visible from across the room for $15. You can do something similar with balance, too. Which is not to say I'm not working on something much nicer and more sophisticated, but if you want a cheap display, there you are!
Their second generation board appears to use 12-bit DACs or ADCs to achieve better resolution than their original design which was based on a vanilla 10-bit capable PIC.
As far as my progress is concerned, yes and no. Every time I send away for a new prototype board my design has changed by the time I it get back in the mail. Today I modified one of my earlier boards in a way which, if it works out, will make a big difference in the long run but also already obsoletes my most recent prototype which arrived on Tuesday. We shall see, I haven't modified my code yet so don't know for sure it'll work but I think it will and I think it'll make a big difference. If it works as I think it will, this is the last major change and things can proceed.
I have spoken to a manufacturer and they've done most of the calculations and are waiting for me to give them a final design so they can give me price quotes for various quantities. The changes I'm making now do not involve much in the way of hardware change except for moving traces and pin assignments so prices should not vary much if at all.
All I can say is stay tuned, and I'm not standing still.
The thing I'm trying to avoid is to start out with something that will not be compatible with the next generation. I want to start with a design that can grow with the times without being obsoleted because it's not compatible. Of course, I'm not a soothsayer and who know where technology will go, but this should be good for a while, anyway.
For the record, let me make clear that I have no direct knowledge or information about these products that you can't glean simply by reading their documentation and looking at the pictures. I could be radically wrong in anything I said, so please take it with a grain of salt -- it's just what I think I understand, not what I know.
More progress
Making some headway now. For several weeks I was immersed in multiple new subjects -- IR control, rotary encoder software, controlling an OLED display, I2C communication -- and it felt like I was getting nowhere with anything. It all began to come together over the past few days, and now I can see forward.
Right now I've got the IR and rotary encoder working together and cooperating via interrupt-driven code, I've got the control chip talking to the OLED display, and I've got the LDR board responding to I2C commands from the controller that will be the heart of the system unless the LDR board is used by itself.
The LDR board can be a stand-alone with pots controlling volume and balance, or under remote control with a choice of IR plus pots or IR plus rotary encoder and display.
The picture is of the breadboard setup -- you can see the rotary encoder knob in the middle and the IR receiver at the extreme front right, the OLED isn't doing much yet except telling the time. But all that stuff is talking to each other and that was the hard part.
Making some headway now. For several weeks I was immersed in multiple new subjects -- IR control, rotary encoder software, controlling an OLED display, I2C communication -- and it felt like I was getting nowhere with anything. It all began to come together over the past few days, and now I can see forward.
Right now I've got the IR and rotary encoder working together and cooperating via interrupt-driven code, I've got the control chip talking to the OLED display, and I've got the LDR board responding to I2C commands from the controller that will be the heart of the system unless the LDR board is used by itself.
The LDR board can be a stand-alone with pots controlling volume and balance, or under remote control with a choice of IR plus pots or IR plus rotary encoder and display.
The picture is of the breadboard setup -- you can see the rotary encoder knob in the middle and the IR receiver at the extreme front right, the OLED isn't doing much yet except telling the time. But all that stuff is talking to each other and that was the hard part.
Attachments
Last edited:
I'm not sure what language you are programming in, but a simple encoder routine written in 8051 assembly is attached. KNOB_I is the input from the encoder that is the steady state. KNOB_Q is the one that pulses when moved and either leads or lags the other in polarity. I and Q are bit locations that store the current values from KNOB_I and KNOB_Q respectively. MIN_VALUE, MAX_VALUE and CURR_VALUE should be self explanatory.
The only other thing you have to do is initialize the value of bit I (from KNOB_I) at power up reset.
The code resides in your main loop, it does not need to be interupt driven (unless your main loop is exceedingly slow).
The only other thing you have to do is initialize the value of bit I (from KNOB_I) at power up reset.
The code resides in your main loop, it does not need to be interupt driven (unless your main loop is exceedingly slow).
Last edited:
Not assembler, but thanks anyway.
And the code for the encoder is already finished and working well.
Late last week the manufacturer who is going to build the boards discovered that there had been a misunderstanding and their PCB supplier can't accept ExpressPCB files after all, so on Friday I started looking around for an alternative solution.
I found a software package that appeared to convert ExpressPCB to Gerber -- its called "Copper Connection" from David Cook's Robot Room: Robotics, Circuits, and Machining. I downloaded the free version which did open my ExpressPCB files and allowed me to edit them but not convert to Gerber. So I bought the paid version and easily converted the LDR board file to Gerber and the files seem to be good, with all the right files in the standard zip package and they looked right when I checked with a Gerber viewer. If the files are indeed good I should have pricing info next week.
I found a software package that appeared to convert ExpressPCB to Gerber -- its called "Copper Connection" from David Cook's Robot Room: Robotics, Circuits, and Machining. I downloaded the free version which did open my ExpressPCB files and allowed me to edit them but not convert to Gerber. So I bought the paid version and easily converted the LDR board file to Gerber and the files seem to be good, with all the right files in the standard zip package and they looked right when I checked with a Gerber viewer. If the files are indeed good I should have pricing info next week.
Now that things are moving into final assembly, is there any indication if the final costing will be comparable to (or better than) the Tortuga units, if you go ahead with production?
I'm a bit surprised that any pcb manufacturer couldn't convert the Express design to any other system for you - you might like to check the Gerber solder masks for correct layer/orientation and drill hole sizes as this seems to be overlooked sometimes
Steady progress now ...
I'm a bit surprised that any pcb manufacturer couldn't convert the Express design to any other system for you - you might like to check the Gerber solder masks for correct layer/orientation and drill hole sizes as this seems to be overlooked sometimes
Steady progress now ...
James, I'd given the BOM and Expresspcb files to the assembler to get pricing for a variety of quantities, and then everything stopped because they couldn't get a price for the board. They order the boards in panels and assemble them that way and then break them apart after the parts are installed. Easier to stuff and wave solder that way I guess.
Regarding price, I don't know what the cost will be, in large part depends on the size of the first run.
Regarding price, I don't know what the cost will be, in large part depends on the size of the first run.
Ah, I was forgetting you were getting them assembled - makes better sense.
The complete set of modules for the Tartuga looks like about $700 - not the cheapest LDR project, but does come assembled, tested and with a guarantee, I guess.
The complete set of modules for the Tartuga looks like about $700 - not the cheapest LDR project, but does come assembled, tested and with a guarantee, I guess.
When jslot asked about rotary encoders things got a little hectic around here. I went back and learned how to use those, plus I2C so various boards could talk to each other, and looked into enabling a display of some kind.
Now that I'm comfortable that I know what the hardware requirements are, I've modified the earlier boards and designed one entirely new board. I now have three boards in the stable -- the original LDR board modified to be controlled by pots or by I2C and the ability to sense automatically if I2C is connected or the local potentiometers are active -- plus the motorized pot board now modified to interconnect with I2C, and a new rotary encoder & digital display board. Both the latter two boards have IR control integrated.
The size of the LDR board has stabilized at 2.5 x 3.5 inches, and the motorized pot board adds 1.25 inches to the 3.5 while the encoder board adds 1.5 inches. Both add-on boards and the LDR board can be used stand-alone for its specific purpose, or the control boards can be screwed to the LDR board via header pins that fit into matching terminals on all the boards or wired and installed separately and connected together with I2C.
The LDR board is ready for a trial run by the manufacturer, and I'll finish verifying the connections on the other two boards and write the software while the main board is getting it's trial production run.
At least, that's the plan.
Now that I'm comfortable that I know what the hardware requirements are, I've modified the earlier boards and designed one entirely new board. I now have three boards in the stable -- the original LDR board modified to be controlled by pots or by I2C and the ability to sense automatically if I2C is connected or the local potentiometers are active -- plus the motorized pot board now modified to interconnect with I2C, and a new rotary encoder & digital display board. Both the latter two boards have IR control integrated.
The size of the LDR board has stabilized at 2.5 x 3.5 inches, and the motorized pot board adds 1.25 inches to the 3.5 while the encoder board adds 1.5 inches. Both add-on boards and the LDR board can be used stand-alone for its specific purpose, or the control boards can be screwed to the LDR board via header pins that fit into matching terminals on all the boards or wired and installed separately and connected together with I2C.
The LDR board is ready for a trial run by the manufacturer, and I'll finish verifying the connections on the other two boards and write the software while the main board is getting it's trial production run.
At least, that's the plan.
Last edited:
Actually, up on the site now is their 3 unbalanced input unit ( LDR3x.V2.1 ) that includes all the modules, a metal box and just about all the required fittings (sockets, knobs, wire, simple remote control, etc) for $695, for a January delivery (a promotion saving of $100)
[This is a kit of modules (Controller brd, Input Relay brd, Rotary Encoder brd, Display brd, Power Supply brd, I/R Receiver brd) that does still need to be assembled, not ready to use 'off the shelf' as I thought.]
There's a couple of things that caught my eye - the LDRs are mounted on small pcbs that plug into header sockets, so are easy to replace, if needed - not a bad idea.
The other thing is they use a method of onboard memory that can be updated at a later stage via a usb connection - not sure how useful this would be in practice, but ...
One thing that might be worth considering (a bit late, I know!) is that there are some devices out there that do actually have some dc voltage on the signal line, usually unknown to the owners, and this is a disaster to the LDRS
I can't think of anyway to protect the units apart from maybe incorporating a dc test on the signal lines at system start up - not sure this is at all feasible in practice without effecting the operation of the LDRs - a really useful safety feature, if it can be done..
[This is a kit of modules (Controller brd, Input Relay brd, Rotary Encoder brd, Display brd, Power Supply brd, I/R Receiver brd) that does still need to be assembled, not ready to use 'off the shelf' as I thought.]
There's a couple of things that caught my eye - the LDRs are mounted on small pcbs that plug into header sockets, so are easy to replace, if needed - not a bad idea.
The other thing is they use a method of onboard memory that can be updated at a later stage via a usb connection - not sure how useful this would be in practice, but ...
One thing that might be worth considering (a bit late, I know!) is that there are some devices out there that do actually have some dc voltage on the signal line, usually unknown to the owners, and this is a disaster to the LDRS
I can't think of anyway to protect the units apart from maybe incorporating a dc test on the signal lines at system start up - not sure this is at all feasible in practice without effecting the operation of the LDRs - a really useful safety feature, if it can be done..
Hmm, a box with all the goodies for $700.
Regarding the plug-in LDRs, yes, I've thought about that and am contemplating doing it at some point. Is LDR failure not uncommon? I guess if you've got DC on the line anything could happen. Does anyone out there have a good handle on how big a problem is LDR failure due to voltage/current on the signal line?
Any test to determine if there is voltage on the line would require a physical connection to the signal line. To do that automatically would require a relay setup, and that could be implemented with the same hardware as an automated calibration system. I don't know if I'm being unreasonable, I just don't like the idea of inserting a set of relay contacts in the signal path. If you're gong to do LDRs, wouldn't that be a step backward? Ditto for using relays for source selection. If the volume control is an LDR, don't you owe it to the signal to use LDRs for source selection too?
Re the USB connection to update the chip -- I can already do that, it's what I do when I update the code on an already-built board with the chip installed. The question I have is security for the code. Once in the chip it would take the NSA to get it back out but updating "in the wild" leaves it rather vulnerable.
- Status
- Not open for further replies.