Lm4702

[gooday]

That being said, if I end up using TO-247 transistors I may make an output board or integrate the output stage into a different design, but I haven't really planned on doing that as of right now. If I do, I will rethink a few aspects of the board to try and protect the signal paths a little better. This weekend I hope to finish up the power supply and check out the stage that is finished with a function generator. I'll be sure to let everyone know how it turns out, but I wish I had all of my transistors to let you know how it sounds!

Good,waiting your good news!

Thanks!
Gooday

 

[gooday]

Woohoo! International mail FINALLY got me my boards from Olimex (the self etched prototype sort of sucked), but still no transistors. Everything else is primed and ready to go when they come in. I have already made several adjustments to the board and am planning on making it a two layer board if I ever order more or want to change components drastically.

PS Sorry for the bad picture. All I have is the camera on my phone, and as I am sure you see, it rather sucks, but you should get the point.

He he.....it's needed to go to buy a digital camera.

 

[dfdye]

I (finally!) have everything but the output stage up and running. With the LM4702 stage wired as per the test circuit in the app sheet, I am getting a nice signal at the sink/source "output" according to my old school scope. I have not hooked up my digital storage scope yet, so I can't post traces yet, but sine wave inputs look really nice at the output. It is at least nice to verify the boards I had made work! Now it is time to put a good load on the board and see how it works.

 

[nicolas888]

Hello

I am working on this amplifier too.

I didn't succeed to make it working due to some oscillations.
I uses different darlington and without a real success

Also, find the SANKEN preconised output transistor is impossible mission.

To review,

Someone could tell me witch output transistor is the best choice ?
Do someone have a diagram ?

I notice that MJ11032 MJ11033 seems to be a good choice.
SAP16 seems to be harder to find.

Do you think that a MJE340 is suitable to replace the SAP16 ?

Thanks

 

[hayenc]

http://www.profusionplc.com/

Got the Sanken SAP16's shipped to US in about a week.

 

[dfdye]

I thought I would share my latest board revision that I did while I was revising my prototype design and trying to decide on what coupling caps to use in the signal path. I couldn't really come up with ONE that I really liked, so I made spots for all of them I was considering in order to let me decide later. As a result, there are spots for a variety sizes of electrolytic bipolar caps (2.5mm, 3.5mm and 5.0mm pin spacing and up to 13mm diameter package)

There is only space for 15mm pin spacing bypass caps, but different sized bypass caps can be accommodated pretty easily.

Also, I took a cue from Russ and others and added in spots for 220uf bypass caps on the power rails. FYI, I used Nichicon Muse KZ 220uf X 100V caps for the size (7.5mm pin spacing X 16mm diameter), but again, I put in space for a different cap (5mm pin spacing) if I go with something else .

All of the signal resistors can accept up to a normal 1/2W package, though the mute resistor will have to be 1/4W.

Also new is the isolation of the signal and power grounds via RG. The flooded ground plane on top is connected to the power ground, and all of the signal grounds are connected to my attempt at a star ground at RG.

I have also alternated input and output signals to the chip on top and bottom layers to separate the signal paths as much as possible.

The overall size is ~2.5in. X ~2.5in.

I am not making an output board at this time since I haven't really gotten a transistor layout that I like yet. I have been playing around with building separate darlingtons with some luck, but of course they are a little more complicated than just throwing in a prepackaged darlington. Still, I am not happy with my output stage, but it is coming along.

I'll post more about the output stage when I get it finished.

 

[dfdye]

I will be ordering a few of these boards in the next couple of weeks. If anyone wants to get in on the purchase and pick up one or more, let me know. This is DEFINITELY a prototype board, and I have tried to keep it flexable to allow for whatever parts you want to throw on the board (within reason, of course!). As far as pricing is concerned, if I can get to 15 parts, Olimex can do them for ~$10 per board by my quick math. Of course they are not the best boards in the universe, and I would like to get somethining better, so if we get more or if you have an idea, let me know and I will see what I can do, but at least we can get the ball rolling without too many people wanting boards.

Whatever the outcome, I figured I would offer to head up a small GB for some boards to get the ball rolling on getting some of these into working amps, if for no other reason to increase the knowledge base about what works well with these chips.

Anyway, if you are interested, let me know and if we get enough people on board I will drop in an order soon.

David

PS If you are interested, please double check the board layout and post suggested changes if you have them. I will be happy to incorporate as many ideas as possible to accomidate as many people's design ideas as I can!

 

[hayenc]

Put me on your list for a couple.

 

[dfdye]

Done!

To keep this thread on topic, I posted a new thread under the group buy forum here

So if you are interested in a board, let me know in that thread, or e-mail me at dfdye at netcape dot net, or click on the e-mail button at the end of the post!

 

[veteran]

I'm watching this topic from a long time . My LM4702 samples are waiting for first tests.

What do you thing about using some low input capacity MOSFETs with this driver?

This is my version of test board for LM4702 . Standard types of darlingtons or MOSFETs can be mounted here (TO3P package). Ground plane in star topology.

An externally hosted image should be here but it was not working when we last tested it.

 

[dfdye]

I'm watching this topic from a long time . My LM4702 samples are waiting for first tests.

Great! I would love to hear how they turn out!

What do you thing about using some low input capacity MOSFETs with this driver?

Not sure, but you should try them out and let us know how it goes!

This is my version of test board for LM4702 . Standard types of darlingtons or MOSFETs can be mounted here (TO3P package). Ground plane in star topology.

I like the layout, but R1 and R12 seem to be hanging. I may just be having a bad image on my monitor (actually I am pretty sure that they are connected after triple checking), but you might want to double check those connections.

Also, I am a little skeptical of having the chip and the output transistors on the same heat sink. Though having the transistors all on one heatsink is obviously necessiary, I would lean toward putting the voltage gain stage away from the high heat devices. I am leaning toward having the output stage on a different boards or wiring them point to point, but the simplicity of a one board amp is appealing.

Thanks for sharing!

PS: You may want to add in spots for bypass caps for the electrolytic input caps. Unless you get ultra high end electrolytics, they will probably benefit by adding in a small film bypass cap to each one.

 

[dfdye]

Quick reminder: if you want in on the group buy for PCBs, I will be placing the order soon, so hury over to that thread and let me know if you want boards!

David

PS If you have any last minute suggestions as to different parts (especially bypass caps for the electrolytic coupling caps) let me know and I will make last minute changes as needed.

 

[gooday]

Quick reminder: if you want in on the group buy for PCBs, I will be placing the order soon, so hury over to that thread and let me know if you want boards!

David

PS If you have any last minute suggestions as to different parts (especially bypass caps for the electrolytic coupling caps) let me know and I will make last minute changes as needed.

Dear Dfdye,

I am in China,it is very trouble to buy your PCB,could you please send your PCB layout to me for making PCB by myself?thank you very much!

Gooday

 

[dfdye]

I honestly don't know! Can you get a pay pal account? I have no idea, but I am sure we can work something out if you can't send money. E-mail me and let's figure something out!

David

 

[veteran]

No dfdye, everything is OK with R1/R12. Transistors can be mounted under pcb so they can be separated from LM4702.

There are no input capacitors on this PCB . Like I said - this is my platform for testing. I have ordered about 20 of them today and can't wait to receive them

An externally hosted image should be here but it was not working when we last tested it.

 

[dfdye]

No dfdye, everything is OK with R1/R12. Transistors can be mounted under pcb so they can be separated from LM4702.

Nice.

What weight copper are you planning on using on the boards?

 

[veteran]

About 100um. Because this board will be through hole plated so there should be something like 110-120um after this process.

 

[jackinnj]

Just sent off for some prototypes -- for use with the Sanken SAP16's -- with or without the input cap (at your pleasure!):

An externally hosted image should be here but it was not working when we last tested it.

 

[hayenc]

Do you have any plans for making the SAP16 boards available for sale?

 

[RCBandwidth]

Lookin Good

Hello Jack: The layout looks great,I am certainly interested in getting a few boards when your ready for production.

I'ts a little odd that those Sanken transistors are not easy to get domestically...

Bob C.