peranders,
might I suggest a circuit similar to the one proposed by Steven here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=28218&perpage=10&pagenumber=1
But with a balanced input, the load (speaker) push-pull coupled, and a way to switch off just the current biasing circuitry.
Just some ideas.
might I suggest a circuit similar to the one proposed by Steven here
http://www.diyaudio.com/forums/showthread.php?s=&threadid=28218&perpage=10&pagenumber=1
But with a balanced input, the load (speaker) push-pull coupled, and a way to switch off just the current biasing circuitry.
Just some ideas.
New specifications:
* 50 um pcb with groundplane
* Rectifier diodes, TO247 and option for TO220
* 4700 uF/50V max, 10 mm snap-in and regular ones too.
* LM3886, inverting, 1k+18k, properly decoupled
* Mute
* Output filter, 1.5 mm wire, 5 mohms, 0.7 uH
* Zobel network
* DC-servo, AD8620, properly decoupled (possible not to use it)
* Input buffer, non-inverting, 220 kohms input impedance, AD8620, properly decoupled.
* LM317/337 in SOT223
* All elctrolytic caps hole mounted
* SMD caps PPS, X7R, NPO types, 0805, 1206, 2220 size
* SMD resistors, 0805
* Lug connectors with option for 2.5 mm2 wires
* Fuses 5 x 20 mm
The goal is a complete product, just add two windings with AC, signal in and speaker out and a cute small pcb.
* 50 um pcb with groundplane
* Rectifier diodes, TO247 and option for TO220
* 4700 uF/50V max, 10 mm snap-in and regular ones too.
* LM3886, inverting, 1k+18k, properly decoupled
* Mute
* Output filter, 1.5 mm wire, 5 mohms, 0.7 uH
* Zobel network
* DC-servo, AD8620, properly decoupled (possible not to use it)
* Input buffer, non-inverting, 220 kohms input impedance, AD8620, properly decoupled.
* LM317/337 in SOT223
* All elctrolytic caps hole mounted
* SMD caps PPS, X7R, NPO types, 0805, 1206, 2220 size
* SMD resistors, 0805
* Lug connectors with option for 2.5 mm2 wires
* Fuses 5 x 20 mm
The goal is a complete product, just add two windings with AC, signal in and speaker out and a cute small pcb.
Ah, I see. I usually place the component symbols overlapping, so that they appear as single in the schematic but make all neccessary pins/pads available on the board. 
Due to the layout software I use, a rough 80mm x 100mm is where I start.
I would expect the components in your schematic to need not more than 3" by 2" including wide supply and gound traces.
But finally, the industrial smd GC will have to compare with the available (non-smd, unbuffered) boards, and those are small already...
I'm looking forward to see your results.
edit: Although I'm making own boards, I'm still interrested in your alpha/beta testing. At last, I want to use the smaller and better boards, not neccessarily the ones that come out first .And I need a lot of them, too...
Due to the layout software I use, a rough 80mm x 100mm is where I start.
I would expect the components in your schematic to need not more than 3" by 2" including wide supply and gound traces.
But finally, the industrial smd GC will have to compare with the available (non-smd, unbuffered) boards, and those are small already...
I'm looking forward to see your results.
edit: Although I'm making own boards, I'm still interrested in your alpha/beta testing. At last, I want to use the smaller and better boards, not neccessarily the ones that come out first .And I need a lot of them, too...
When I have alternative parts I can do it in two ways, Make more holes in one part or just add the second one. The advantage of the later is that you will get the shape (and polarity, important when it comes to diodes) of if and also get those in the parts lists. My software has a totally integrated database.
If you feel that you are up to building and testing drop me a note when I have the boards.
If you feel that you are up to building and testing drop me a note when I have the boards.
i find that the 3-pad "SOT" small signal diodes are easier to place than the 2-pad variety -- probably doesnt matter that much as they are just there to protect the servo and would probably be mounted correctly -- the 3-pad variety would assure that the protection diodes are oriented correctly.
it looks like a very nice project
it looks like a very nice project
The pcb is now ready and can be downloaded from here
http://home5.swipnet.se/~w-50719/hifi/qrp02/downloads_qrp02.html
http://home5.swipnet.se/~w-50719/hifi/qrp02/downloads_qrp02.html
An externally hosted image should be here but it was not working when we last tested it.
The silk print
An externally hosted image should be here but it was not working when we last tested it.
Component side
An externally hosted image should be here but it was not working when we last tested it.
Solder side
An externally hosted image should be here but it was not working when we last tested it.
Ground plane on solderside. Power ground to the left and signal ground to the right. The rectifier bridges don't have a ground plane.
Size 4" wide and 4.3" high
I have written a circuit description see link. Remember that this page is in a frame (no navigation).
http://home.swipnet.se/~w-50719/hifi/qrp02/schema_qrp02.html
http://home.swipnet.se/~w-50719/hifi/qrp02/schema_qrp02.html
peranders said:
Per: one of the folks at Stereophile modeled the RLC of a real-world speaker with cables -- I used this when modeling the GC's that I use -- I found that the only effect of the resistor value was to change the Q of the coil.
peranders said:
From this worm's eye-view the PCB's look good -- have you checked them with a Gerber-Viewer ? I doubt there are any errors, but it's worth the time investment.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.