My LME49600/49610 PCB design

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi All

Attached is a pic of my 4-layer multi-channel LME49600/49610 PCB design originally inspired by diyAudio Member OPC.

I originally designed the PCB for a 10 channel mini surround sound system but I also added provisions for multiple channel configurations. The PCB can be configured using solder jumpers underneath the board to operate in mono using up to 24 LME49600/LME49610 devices in parallel or stereo with a maximum of 12 devices in parallel per channel to twelve separate channels with a pair of LME's per channel and everything in between. The input op-amps can be paralleled for additional drive and noise reduction when configured in mono or stereo. There's also DC speaker protection on-board for each individual channel. The circuit also prevents speaker thumping when power is turned on and off. The relays can bypassed or paralleled using solder jumpers to lower relay contact resistance if desired.

I've yet to have the PCB manufactured but If there's enough interest from members of this great forum I'd be more than happy to have the PCB manufactured tested and made available.

I also have a PS to power it. Two PS are required one for each side.

Best Regards,

Al
 

Attachments

  • LME49600_PCB.jpg
    LME49600_PCB.jpg
    274.4 KB · Views: 773
Last edited:
aha sorry yes its the optional pads for paralleling I was seeing as a place for current sharing resistors. what sort of speakers were you thinking to use for it in multichannel mode? very efficient ones I guess. interesting design anyway, I would have liked to see the psu on the same board. all the same since I have a fair few left over buffers I may grab one of these.

what topology is the supply board you have planned? are you also using lme49990 for the VAS? as this is pretty low current, to keep things neat, you could look at using an optional DSUB25 connector for the power umbilical and output
 
Last edited:
Hi qusp,

For the multichannel mode the small speakers I'm using are actually not that efficient at about 82db/1watt/1m. But they don't need to be since the speakers are positioned relatively close to my ears.

I kept the PS external for added flexibility to select your own PS which seems pretty popular these days. If it was mono like OPC's design I'd go on-board for sure like I did with my version of SymAsym. Also the PCB measures 7.6" x 3.6", adding the PS on-board would of made the board quite a bit larger adding to the overall cost. I'm also a stickler for neatness, one or two on-board PS what of really cluttered things up.

For the PS I'm using the same topology that OPC used on his design. It's simple enough and proven so I built by own PCB pictured below. I'll being using a 25VA transformer not 15VA as pictured. For added current I'll use LT1085/1033 3 amp regulators instead of the LM317/337.

I'm also using the LME49990 for the front end.

Thanks for your idea of using DSUB25 connector, definitely something to think about.

Al
 

Attachments

  • CIMG1952.JPG
    CIMG1952.JPG
    245.6 KB · Views: 635
Last edited:
Hi qusp,

For the multichannel mode the small speakers I'm using are actually not that efficient at about 82db/1watt/1m. But they don't need to be since the speakers are positioned relatively close to my ears.

I kept the PS external for added flexibility to select your own PS which seems pretty popular these days. If it was mono like OPC's design I'd go on-board for sure like I did with my version of SymAsym. Also the PCB measures 7.6" x 3.6", adding the PS on-board would of made the board quite a bit larger adding to the overall cost. I'm also a stickler for neatness, one or two on-board PS what of really cluttered things up.

For the PS I'm using the same topology that OPC used on his design. It's simple enough and proven so I built by own PCB pictured below. I'll being using a 25VA transformer not 15VA as pictured. For added current I'll use LT1085/1033 3 amp regulators instead of the LM317/337.

I'm also using the LME49990 for the front end.

Thanks for your idea of using DSUB25 connector, definitely something to think about.

Al

I think i've seen that PSU layout somewhere else ;)

I'm thinking of designing an amp using two LME49990 bridged front end driving the LME49600 buffers driving some nice output devices! 18V rails. Ultra low distortion about 50W 8R perhaps? Tweeter amp? All SMD perhaps?
 
yeah look with an identical PSU right down to pretty much the exact layout, identical parts in the front end, this whole thing is cutting a bit close to the bone IMO, consider me uninterested.

publishing a thread asking about doing a run of PCBs under these conditions, linking to the front page and measurements, its really not cool IMO. its one thing to take inspiration from someone, given the performance of this amp its understandable; but to just transpose the whole thing with the addition of a servo and talk about doing a GB on the PCBs, while Owen is trying to make a go of the amp as a product?.... is going too far.

pretty sure you dont have bad intentions, but did you really think this through?
 
Last edited:
The PS supply is not what makes OPC's design unique, Its the amp section and the overall design. There is nothing special in the PS. The difference in his PS than what I've done in the past is that he used SMDs underneath the board to make it more compact but that's about it. The LM317/337 circuit that he used can be found in the datasheets and has been implemented countless times. I seriously doubt OPC will mind, but if so I wont make it available to those that may be interested in my multi-channel PCB which is completely different than OPC's mono design.
 
Last edited:
Sorry Boscoe, but I don't think I did anything morally wrong by making the PS. And qusp I certainly didn't have any bad intentions. It would of been morally wrong had I copied his design exactly and tried to sell the boards but I didn't do that. I designed by own multi-channel PCB based on the LME49600 for a specific application that's currently on hold and I thought maybe some members of this forum would be interested in it. That was my only intention. My apologies if I offended any members.

Al
 
qusp you wrote.

publishing a thread asking about doing a run of PCBs under these conditions, linking to the front page and measurements, its really not cool IMO. its one thing to take inspiration from someone, given the performance of this amp its understandable; but to just transpose the whole thing with the addition of a servo and talk about doing a GB on the PCBs, while Owen is trying to make a go of the amp as a product?.... is going too far.

I majority of all the designs on this forum have been inspired by others. It's what drives innovation. To suggest that I should not have started this thread for my multi-channel PCB design which is completely different than OPC's mono PCB design because OPC is trying to make a go with his product is ridiculous. Should I have to wait for OPC? What if OPC's design never goes commercial. What then?
 
Last edited:
there is a big difference between taking inspiration from and copying the entire circuit part for part including the layout for some of it and then offering a GB; thats the operative part right there, offering it for sale. if you were just making it for yourself, I would think it a bit uninspired, but that would be the end of it, can you really not see this distinction?

there is no chance of it not happening, most everything is ordered AFAIK, the only thing that might stop it at this point would be someone undermining it.... sounds like you didnt even email him.

just how do you think it would go down if you did what you did here to nelson pass and posted the design in the pass forum?
 
Last edited:
Sorry but I didn't say I wanted to make money off the PS or the amp for that matter. And BTW it may look the same but not a single device on the PS pictured is the same except the rectifier diodes. The transformer, the regulators, the 4 tops side caps, bottom caps, variable resistors are all different values. The heatsinks and layout traces are different, and it's only a two layer board not three like OPC's design.

If you think my amplifier is like OPC because I paralleled a bunch of LMEs, really what's so special about that? It's not the first time someone parallels several LME49600. As far as the front end, my design is only SE and doesn't provide balance like OPC's, and the SE circuit can be found in the datasheets.

I created the thread because i felt there was an obvious and clear distinction between the two designs, mine being multi-channel, and OPC being mono. Can't get much different than that. Anyway, I know regret ever posting my design. Geez what a mess.
 
AAK, could you post a schematic of your design? It'd be easier to understand your implementation that way.

I think you haven't intentionally written it this way but there was some implication in the first post that the design in this thread will have measurement like the LPUHP, which isn't necessarily the case since Owen has measured his specific layout and BoM parts choice.

It's surely hard to originate in a bipolar lm3xx PSU.

My personal opinion is that you'll want more power than what you'll get from this as a surround amp even near field.
 
Last edited:
changing brand of parts doesnt mean anything really and certainly wouldnt mean anything to a lawyer if it came to that (unlikely). agreed these are all fairly standard logical ways of putting the parts together each circuit in isolation. the overall layout goes it doesnt touch opcs, but:

• the size of the decoupling caps
• the way they are offset on the buffers pins
• how the buffers are arranged against each other
• the choice of, LAYOUT and arrangement of the PSU
• the exact same active components chosen
• linking and referencing his design and measurements in the first post.
• offering it for sale

youve added a servo, thats the only thing youve changed and it shouldnt even be needed. I get less than 6mV DC at full output on one amp and less than 3 on the other, both numbers I wouldnt trust at that low end of the scale with the meter I used and are likely a result of very small input DC. how much DC people get on the output of these LME buffers seems directly linked to the layout, i've often seen it claimed they have issues with DC, but i've never actually experienced it and I have 5 amps using them currently.

in this case legally I doubt hed have much to go with even if he wanted to, but morally I think the lines are pretty clearly drawn
 
Last edited:
Hi hochopeper,

I currently don't have a complete schematic, but there's really not much to the amplifier circuit. All you have to do is look at Fig. 2 of the LME49600 datasheets for my SE circuit with the LME49600 as an output. The non-inverting input in my design has 1k to ground. It's basically the same design OPC is using for his SE input configuration. He does have a balanced circuit on the input which is not in the datasheets I believe. Other than that my design has optional solder jumpers in order parallel the outputs, and the same for the input op-amps for added drive. A DC speaker protection circuit is on-board that also eliminates turn on and off thumping.

I don't see in my first post where I implied that my design measures like OPC's LPUHP. But my PCB is four layers and carefully layed out for the highest quality performance, but since I have not had the board manufactured or tested I can't say for sure.

When I said the speakers are relatively close to my ears I meant inches. Two pairs of LME49600 in parallel is actually more than enough.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.