I ordered 30 boards for my LM4702 amplifier. This is a test version pcb to take a closer look to National new audio driver. I don't need about 15 of them so if You are interested - just let me know. One board will cost about 12$ - 150x60mm, through hole plated, double layer, about 100um copper.
I would like to start tests from some cheap output transistors like 2SD2390/2SB1560 or 2SD2560/2SB1647 and 2SK1529/2SJ200 mosfets. SAP16N/SAP16P are very interesting but they are hard to buy here.
Documentation:
amp sch
I would like to start tests from some cheap output transistors like 2SD2390/2SB1560 or 2SD2560/2SB1647 and 2SK1529/2SJ200 mosfets. SAP16N/SAP16P are very interesting but they are hard to buy here.
Documentation:
amp sch
An externally hosted image should be here but it was not working when we last tested it.
This was a test board. Capacitors protect speakers but amplifier sound much better without them 
Supply voltage is limited to maximum OPA552 voltage range or (if regulators will be mounted) to about (+)(-)40VDC.
No, not yet. I'm waiting for my boards to start some real tests.
GB is closed. Soon I will put some pictures of my amplifier.
Supply voltage is limited to maximum OPA552 voltage range or (if regulators will be mounted) to about (+)(-)40VDC. No, not yet. I'm waiting for my boards to start some real tests.
GB is closed. Soon I will put some pictures of my amplifier.
Re: lm4702
Unlike other opamp-based amps where there are only a few components present to impact sound, my bet is that every builder's choice of output transistors and configurations will substantially change the character of the amp, even though the initial voltage driver will be the same. Still no proof of this yet, but I'm working on it!
For instance, I already have 3 prototype output stages that I am tweaking, though I haven't gotten them to the point of final testing yet. My guess is that each will have a different character.
What I REALLY like about this chip so far is how flexable it is with regard to output! As long as your current gain stage is solid and can stand the low current, it seems as if the chip is feeding it pretty clean signals. Obviously 5.5mA isn't a lot, but it is definitely enough current to work with.
The potential is for a much higher powered amp going into harder-to-drive speakers since the LM4702 is merely the voltage gain stage rather than the power output device. The overall sound quality of the amp will be highly dependent on the type of output stage you choose, though the initial tests of the chip indicate it is a nice, clean voltage gain stage.steven344 said:
Unlike other opamp-based amps where there are only a few components present to impact sound, my bet is that every builder's choice of output transistors and configurations will substantially change the character of the amp, even though the initial voltage driver will be the same. Still no proof of this yet, but I'm working on it!
For instance, I already have 3 prototype output stages that I am tweaking, though I haven't gotten them to the point of final testing yet. My guess is that each will have a different character.
What I REALLY like about this chip so far is how flexable it is with regard to output! As long as your current gain stage is solid and can stand the low current, it seems as if the chip is feeding it pretty clean signals. Obviously 5.5mA isn't a lot, but it is definitely enough current to work with.
Re: Info
Yes, board will have silkscreen and green soldermask just like that one:
This GB is closed so no more boards available. I will probably make some more in a future.
Dfdye is right. Output stage is most delicate part of this amplifier so we have to find the best way to make full use of 5mA output current from LM4702
dj_vulture said:
Yes, board will have silkscreen and green soldermask just like that one:
An externally hosted image should be here but it was not working when we last tested it.
This GB is closed so no more boards available. I will probably make some more in a future.
Dfdye is right. Output stage is most delicate part of this amplifier so we have to find the best way to make full use of 5mA output current from LM4702
alexcd said:
I think that I will have few additional sets. I have confirmation that boards are now ready and probably I will receive them Friday.
There is a new application note from National with few more details about LM4702 and amplifier basing on it:
LM4702 amplifier application note
I think that my board is quite similar to project proposed by National
I can't believe I missed this!!! Thanks for the link.veteran said:
I think this note is probably a pretty good reflection of what the DIY'ers have figured out about this chip too!
The one thing that I really noticed about their layout vs. all of ours is their IC is mounted on the center of the board. We have all been moving ours toward the edges of the board assuming some pretty good sized heat sinks for the chip. It looks like they didn't even sink thier's!
veteran said:
There is a new application note from National with few more details about LM4702 and amplifier basing on it:
LM4702 amplifier application note
100W per channel....not that much. Can it be more? Somethink like 200W/channel at 4 Ohm?
Absolutely! That's why most of us are concentrating on building different output stages than simple darlingtons. This particular board is limited to darlingtons, but the chip itself can use a much more robust output stage and crank some serious power based on the max rail voltages. (+/- 75V for the "C"'s and +/-85 for the A's and B's, though the latter two aren't out yet). If you do the math, you can get some serious crank out of those. The trick is the right output configuration. The tradeoff, obviously, is a more complicated and less straightforward (read more expensive) amplifier circuit.
David
David
To reply to my own comment, I re-checked the National specs, and the chip can do just fine without a heat sink. Even National agrees: http://www.national.com/kbase/category/Audio.html#33dfdye said:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.