I disagree with all of that, but fail to see how this is relevant in this thread so lets agree to disagree and get back to the topic.
Just curious, how are you going to solder the shielded SMD inductor to the board? I think typically they are soldered by "reflow" using solder paste and oven. Toriodal type inductors can be readily soldered to the board but then they are not shielded.
Regards,
If you're going to use two amps per side anyway, might as well parallel the FR amp too. I can't see how this might be of any harm.
That's the obvious choice isn't it? The only problem is that I may want to experiment with a tweeter at some point (probably not, but you know, maybe) and the extra channel would be useful if I do. Removing a dummy load would be simpler then, than "unparalleling" the amp. That said maybe the procedure for going between PBTL and BTL on the green amp is so simple as to not make a difference. I should check into that one.
😀
I hand't quite got round to working that one out yet. I was already realising that I'll need to temporarily remove some of the caps around the current inductors to get access to them so that I can desolder them from the board.
Might have to go for toroidals instead as I don't think the wife would let me use the oven for solder reflow 😉.
Last edited:
Better not ask and wait until she leaves the house then 😛
Really though, it's pretty much a harmless procedure (to the oven that is) so why not?
I think you can also just solder the exposed edge to the pad with conventional soldering iron. There is plenty of contact room for sufficient current flow. I have done it before with the 0.5 in square SMT inductors before. Works fine.
Also possible is to attach some wire lead 'legs' to the SMT component and solder it as conventional thru hole.
Also possible is to attach some wire lead 'legs' to the SMT component and solder it as conventional thru hole.
Last edited:
That was going to be the approach I would take, just wasn't sure if it was feasible.
I might consider doing this to make it easier to remove/change them at a later date.
To be honest I think the trickiest part will be removing the existing inductors without destroying the pads.
You guys using more than one stereo amp in your setups, how are you doing with regards to the PSU? Are you using one big PSU for several boards or one PSU per board and what criteria did you use to choose?
I will be using switching PSUs and I'm leaning toward Meanwell PSUs. Possibly 12V for the full ranges and 24V for the basses.
Considering my full rangers are 8 Ohms and won't be drawing that much power, I'm thinking a 12V 2.1A (~25W) PSU may be enough.
For the basses 24V and 4.5A (~100W) should be good. I'll probably be running at the lowest voltage I can adjust them evenly down to (~21V I guess) which should limit my maximum power to around what the speakers can handle RMS iirc and be well within safe limits, while still having enough power and sounding good at decent levels.
I will be using switching PSUs and I'm leaning toward Meanwell PSUs. Possibly 12V for the full ranges and 24V for the basses.
Considering my full rangers are 8 Ohms and won't be drawing that much power, I'm thinking a 12V 2.1A (~25W) PSU may be enough.
For the basses 24V and 4.5A (~100W) should be good. I'll probably be running at the lowest voltage I can adjust them evenly down to (~21V I guess) which should limit my maximum power to around what the speakers can handle RMS iirc and be well within safe limits, while still having enough power and sounding good at decent levels.
PMental,
I have kind of low power requirements as I can't play very loud without getting complaints from SWMBO - about 85dB. I have 85 dB to 91dB efficient speakers and I find that a single laptop brick 19v at 3.5 amps (paralleled to both boards) is plenty sufficient to power both amp boards. I used Pano's "Measure your speakers voltage" test CD tracks which have -12dB (absolute line level) and measured my speakers as needing 8 watts/channel for bass and 4 watts per channel for treble. When SWMBO wasn't around I put on some club dance tracks with a good bass beat and cranked it up. That little laptop brick and dual tpa3116's could still shake the room on bass notes and was uncomfortably loud.
For those of you who want to wire more SMPS laptop bricks in parallel with diodes in between each supply to get more current and power - it may be possible if you use identical brand and model units - I actually started a thread asking this question some time ago. There seemed to be consensus as to trying this but I have not had a chance to try yet as I do not have two identical amps bricks. Although maybe I should just try it with non identical bricks?
If not using smaller laptop bricks, the high current Meanwell supply with 7amps may be a good call and wire everything in parallel. I recall 21v being optimal for best sound quality so you may want to trim the 24v supply down a bit.
I have kind of low power requirements as I can't play very loud without getting complaints from SWMBO - about 85dB. I have 85 dB to 91dB efficient speakers and I find that a single laptop brick 19v at 3.5 amps (paralleled to both boards) is plenty sufficient to power both amp boards. I used Pano's "Measure your speakers voltage" test CD tracks which have -12dB (absolute line level) and measured my speakers as needing 8 watts/channel for bass and 4 watts per channel for treble. When SWMBO wasn't around I put on some club dance tracks with a good bass beat and cranked it up. That little laptop brick and dual tpa3116's could still shake the room on bass notes and was uncomfortably loud.
For those of you who want to wire more SMPS laptop bricks in parallel with diodes in between each supply to get more current and power - it may be possible if you use identical brand and model units - I actually started a thread asking this question some time ago. There seemed to be consensus as to trying this but I have not had a chance to try yet as I do not have two identical amps bricks. Although maybe I should just try it with non identical bricks?
If not using smaller laptop bricks, the high current Meanwell supply with 7amps may be a good call and wire everything in parallel. I recall 21v being optimal for best sound quality so you may want to trim the 24v supply down a bit.
Rhing,
I saw your posts on the TDA7297 on another forum - you seem to have great experience with both amps. What is your assessment of the sound quality and differences between the 3116 and "lunch money" class AB amp? Go ahead and compare them after you have tricked both up with huge supply rail caps and fancy input caps.
Thanks,
X
I saw your posts on the TDA7297 on another forum - you seem to have great experience with both amps. What is your assessment of the sound quality and differences between the 3116 and "lunch money" class AB amp? Go ahead and compare them after you have tricked both up with huge supply rail caps and fancy input caps.
Thanks,
X
PMental,
From what I gathered from this thread, if you're using one common power supply to supply two boards that are close to each other, you would want to sync boards between each other to avoid "hetrodyning"
From what I gathered from this thread, if you're using one common power supply to supply two boards that are close to each other, you would want to sync boards between each other to avoid "hetrodyning"
Hi guys,
If you are paralleling supplies isolated by diodes, should make sure that they are in balance, so that they share the load I as equal as you can get them. I doubt that many bricks even of the same type are the same output V (what's the tolerance of them? 5-10%). Don't see any means to adjust/fine-tune the outputs on a lap-top brick. You mentioned that the Meanwell supply has a trim, so that is a good choice.
You would be better off just using a separate brick for each module or two if you have no means to balance the load. The modules are not required to share a common supply.
I still don't buy this 21V sweat-spot testament, use as much headroom as the system allows, (26V) but everyone is allowed their opinions! Now if you show me a FFT plot showing a difference, I could be swayed 🙂
If you are paralleling supplies isolated by diodes, should make sure that they are in balance, so that they share the load I as equal as you can get them. I doubt that many bricks even of the same type are the same output V (what's the tolerance of them? 5-10%). Don't see any means to adjust/fine-tune the outputs on a lap-top brick. You mentioned that the Meanwell supply has a trim, so that is a good choice.
You would be better off just using a separate brick for each module or two if you have no means to balance the load. The modules are not required to share a common supply.
I still don't buy this 21V sweat-spot testament, use as much headroom as the system allows, (26V) but everyone is allowed their opinions! Now if you show me a FFT plot showing a difference, I could be swayed 🙂
X: I have no limits but my ears and my system when it comes to levels so I need more power than you whipped boys do 😉
Isn't there a risk of some kind of modulation passing through the PSU from one amp to the other when using a single PSU? Say I'm listening to some bass heavy music that is taxing the bridged amp, can that not affect the full range amp? Apart from different power needs that would be a reason to keep them separated I'm thinking. We're not talking huge amounts of money and if this sounds good I may well live with this system for quite a while, so I might as well do it right from the start!
Isn't there a risk of some kind of modulation passing through the PSU from one amp to the other when using a single PSU? Say I'm listening to some bass heavy music that is taxing the bridged amp, can that not affect the full range amp? Apart from different power needs that would be a reason to keep them separated I'm thinking. We're not talking huge amounts of money and if this sounds good I may well live with this system for quite a while, so I might as well do it right from the start!
Have to do the cross modulation test to be sure.
Run a test signal full load on aggressor and measure the test signal on other channels, having no input signal.
I also ran a quick listening test using a var. linear supply, using a TPA3100D2 design, I can not make out any difference at 21V-26V
Last edited:
He, several replies while I was typing mine 🙂
Now that I think about it, to get as much power as possible out of the amps with an 8Ohm load I should probably go with 24V for the full rangers too, shouldn't I? My full range speakers are the Fountek FR135EX: Products_Fountek Electronics Co.,Ltd which will handle 40W RMS and even at 24V I won't get there at 8Ohms.
One 24V 1.5A (~35W) for the full rangers and one 24V 4.5A (~100W) for the basses, does that look reasonable for max available power at 8Ohms and 2Ohms respectively?
EDIT: My reasons for going to 21V on the basses was more to limit power in such a way that I couldn't blow my speakers even at max volume. You know, for those occasions when you come home at 2.30AM and sit down with a glass of single malt and play music at concert levels 🙂
Thanks for the input guys! I think I will go for separate supplies to avoid any potential problems. Like I said it's not a huge money difference.
Now that I think about it, to get as much power as possible out of the amps with an 8Ohm load I should probably go with 24V for the full rangers too, shouldn't I? My full range speakers are the Fountek FR135EX: Products_Fountek Electronics Co.,Ltd which will handle 40W RMS and even at 24V I won't get there at 8Ohms.
One 24V 1.5A (~35W) for the full rangers and one 24V 4.5A (~100W) for the basses, does that look reasonable for max available power at 8Ohms and 2Ohms respectively?
EDIT: My reasons for going to 21V on the basses was more to limit power in such a way that I couldn't blow my speakers even at max volume. You know, for those occasions when you come home at 2.30AM and sit down with a glass of single malt and play music at concert levels 🙂
Last edited: