Rav,
you are misguided.
I think that Daniel's explanation is rubbish.
How can Spike be over-ruled by enclosing the chip amp inside the composite?
Thanks Andrew, that's the picture I was seeking. Anything that comprehensive would, in my limited understanding, remove the result from the MyRef concept. That's perfectly fine if the sound produced is high quality, and I for one would love to participate in formalizing an LME498xx based amp with the features promoted by Daniel. Hope it develops.
Any error at the speaker output is reduced by same proportion as the added global negative feedback loop that goes back to the small signal amp. Otherwise, the nesting concept would be completely useless.
Daniel was complaining that Spike protection is the 3886 problem.
Daniel went on to try to say that enclosing the 3886 inside the Howland and inside the composite that the combined arrangement reduced or eliminated the undesirable effects of Spike protection.
That's what I read and understood from post115.
Have I got it wrong, or is Daniel back on Venus?
Daniel went on to try to say that enclosing the 3886 inside the Howland and inside the composite that the combined arrangement reduced or eliminated the undesirable effects of Spike protection.
That's what I read and understood from post115.
Have I got it wrong, or is Daniel back on Venus?
That's true (in proportion to the gain).
With nesting, errors at the output are merely reduced, not eliminated.
But, any port in a storm, right?
P.S.
I'm not offering nesting as an excuse to put a huge amount of effort into a TV chip that contains a hard limiter. Personally, I'd rather have an LME chip (driver chip), or one of ST's with the softie limiter. But, then I wonder if when using one of ST's chips if I'd like to have the quality control assistance of nesting? And I'd just have to answer Yes.
Last edited:
Any port in a storm is just like the Fix-A-Flat product that makes the flat tire round and puts in some glue. That's not the final answer, but it was much better than a flat. In context, I said that nesting basic quality active parts is an improvement, but not a substitute for high quality active parts. That resolves: Nested LM3886 is not a substitute for LME driver chip.
Last edited:
Fix-A-Flat stuff, oh ya we call it Elephant snot. Yes, I agree LME drivers are nice parts, so back to working on my LM49830/LatFet design, any advise for me on this?
http://www.diyaudio.com/forums/solid-state/220962-lme49830-latfet-yet-another-design-review.html
Thx Rick
http://www.diyaudio.com/forums/solid-state/220962-lme49830-latfet-yet-another-design-review.html
Thx Rick
Actually yes. Your design requires a DC servo.
The wrong NFB cap value is shown and that will make for loud mids with boomy bass sound. However, the right value is so large (1000u~2200u) that it would "Taser" damage the input (or cause bouncing offset if you tried bypass diodes). Impasse. That design has no correct value of NFB cap. Therefore it needs a DC servo.
Operating voltage at 100v+100v (???), but gain of 28, means there's no source device for it--I think you need to include a preamp run on discrete regulators. The design shown has enough power; however, I doubt you can find enough speaker.
Perhaps a bit lower voltage to reduce extra costs at heatsink and transformer?
And, if you should happen to get a forwards sound for that power amp, you can copypasta this from my TDA7294 power circuit: "Pair 220u amplifier board power caps and a single 2u polyester rail to rail." I'm just saying that while you're paying shipping for those 100u, you might also want to purchase some 220u to have the option available.
The wrong NFB cap value is shown and that will make for loud mids with boomy bass sound. However, the right value is so large (1000u~2200u) that it would "Taser" damage the input (or cause bouncing offset if you tried bypass diodes). Impasse. That design has no correct value of NFB cap. Therefore it needs a DC servo.
Operating voltage at 100v+100v (???), but gain of 28, means there's no source device for it--I think you need to include a preamp run on discrete regulators. The design shown has enough power; however, I doubt you can find enough speaker.
Perhaps a bit lower voltage to reduce extra costs at heatsink and transformer?And, if you should happen to get a forwards sound for that power amp, you can copypasta this from my TDA7294 power circuit: "Pair 220u amplifier board power caps and a single 2u polyester rail to rail." I'm just saying that while you're paying shipping for those 100u, you might also want to purchase some 220u to have the option available.
Last edited:
I've used the LM3886 in several projects and am happy with the sound. These amps are a commodity and pre-built pcbs can be purchased on Ebay, or you can build up one of several high quality versions (e.g. from ChipAmp.com, etc.). But there is another tact that I have been on lately...
I've started using some old Sanyo STK chip amps, the series with STK41x1V where 1<x<9 (e.g. 4111V through STK4191V). I like these because... they are CHEAP and EASY to put to use! Performance is reasonable and distortion for the most part low enough to be much less than the distortion from the loudspeaker drivers themselves. I build active speakers, so I need to put several chip amps and their power supplies inside a speaker box. I recently started doing some distortion testing using the spectrum analyzer offered as part of the free "Room EQ Wizard" software. This has helped me investigate and solve some problems with grounding and hum, as well as to get more intimately familiar with the distortion profile of the Sanyo chip amp. The main weakness of these chip amps comes above 3k-5k Hz, where distortion starts to climb, sometime above 0.1% THD. Otherwise distortion is less than 0.05% typically and can be less for low frequencies. 2nd order dominates for all frequencies.
These chipamps are dual types, like the LM4780, and the 25W+25W STK4141V can be purchased as part of a plate amp including the pcb, heat sink, on-off switch, line cord, etc. from a couple of sellers in the USA. I make several deletions and modifications, which takes a little time, but I still come out ahead. I recently started swapping out the STK4141V for the STK4191V, which has 50W+50W power rating into 8 ohms and can easily run 4 ohms loads. The STK4191V can actually drive 8 ohms loads to more than 100W+100W but this would require (at a minimum) a better pcb with heavier tracks and a large heatsink.
I have used these chip amps bridged and as a two channel amp in active loudspeaker projects. For better high frequency performance, I would use the LM3886 which has much lower distortion above 5k Hz and can easily drive 4 ohms loads (like just about every good tweeter out there) unlike some other popular chip amps from National.
I thought that my decidedly different take on chip amps would be something to throw into the mix here...
-Charlie
I've started using some old Sanyo STK chip amps, the series with STK41x1V where 1<x<9 (e.g. 4111V through STK4191V). I like these because... they are CHEAP and EASY to put to use! Performance is reasonable and distortion for the most part low enough to be much less than the distortion from the loudspeaker drivers themselves. I build active speakers, so I need to put several chip amps and their power supplies inside a speaker box. I recently started doing some distortion testing using the spectrum analyzer offered as part of the free "Room EQ Wizard" software. This has helped me investigate and solve some problems with grounding and hum, as well as to get more intimately familiar with the distortion profile of the Sanyo chip amp. The main weakness of these chip amps comes above 3k-5k Hz, where distortion starts to climb, sometime above 0.1% THD. Otherwise distortion is less than 0.05% typically and can be less for low frequencies. 2nd order dominates for all frequencies.
These chipamps are dual types, like the LM4780, and the 25W+25W STK4141V can be purchased as part of a plate amp including the pcb, heat sink, on-off switch, line cord, etc. from a couple of sellers in the USA. I make several deletions and modifications, which takes a little time, but I still come out ahead. I recently started swapping out the STK4141V for the STK4191V, which has 50W+50W power rating into 8 ohms and can easily run 4 ohms loads. The STK4191V can actually drive 8 ohms loads to more than 100W+100W but this would require (at a minimum) a better pcb with heavier tracks and a large heatsink.
I have used these chip amps bridged and as a two channel amp in active loudspeaker projects. For better high frequency performance, I would use the LM3886 which has much lower distortion above 5k Hz and can easily drive 4 ohms loads (like just about every good tweeter out there) unlike some other popular chip amps from National.
I thought that my decidedly different take on chip amps would be something to throw into the mix here...
-Charlie
LM3886's hard limiter sounds really great for clear bass and bari. ST's softie current limiter trims bass but doesn't hurt treble. For Bi-Amp, I'd prefer LM3886 on bass, TDA7294 on treble. Have your cake and eat it too.
The amplifier board power caps size reflects the size of the signal, such as 2200u for LM3886 bass amp, 220u for TDA7294 treble (or full bandwidth) amp. I'd run the TDA7294 non-inverting; however, I'd run the LM3886 inverting t-network (see Decibel Dungeon) for the cooler temperatures of solid stability that may relate to lower heatsink expense.
You shouldn't need the chip amp limiting/protection circuitry to kick in, so this should not be part of the argument for using the chip amp... don't overdrive your chip amp and you won't need to worry about it!
The LM3886 has 10 times better distortion performance above 2k Hz than the TDA7293 or TDA7294. The TDA's can be run at higher rail voltages. I would use the TDA's (possibly multiple TDA7293s in parallel using master/slave configuration to raise current capability) and then use a single LM3886 on the tweeter.
-Charlie
Yes, he does have a good reputation.OK Daniel, you have peaked my curiosity beyond my resistance level
The board is mostly compatible with post #98
Consider upgrade/retrofit the board to signal star ground.
For treble amp, you want a FAR smaller input cap.
Why not point to point?: I prefer Authentic chip, Omit PCB distortion. For build method, click the LM1875 parallel in my signature line, since point to point TDA7294 build method is visually similar (move the power pins up away from signal pins). For point to point TDA729x, the mute just uses 10k, but the standby uses 22k delayed by 10u.
Point to point for TDA7294 wasn't quite as easy as I thought, but here it is:
http://www.diyaudio.com/forums/chip...t-no-pcb-tda7293-tda7294-tda7295-tda7296.html
P.S.
You might want to buy the boards because it will be much easier to swap parts on the board.
A point to point build can be done after you're already sure of what parts you'd like to put on it.
Not much data, but any thoughts on this one?:
Finished TDA7293 Hi Fi Audio Power Amplifier with Speaker Protection? | eBay
Finished TDA7293 Hi Fi Audio Power Amplifier with Speaker Protection? | eBay
Needs heat spreader bar, air vent holes and at least 1/2cm tall feet all added to the bottom of the case, pronto! Needs the bootstrap and NFB caps check to assure that they're at least up to minimum required values for quality bass, 47u min for bootstrap 220u min for NFB cap, and although specific values would be better, those minimums are more suitable than the datasheet's 22u for both. The power circuit needs checked for quality, and I, personally don't favor all-in-one integrated power supply + amplifier all-in-one board with TDA729X chips.
However, the materials used in the construction of that amp cost almost as much as its price, so it is actually a fairly good value, and you wouldn't have to do a lot of annoying casework by hand, since the stuff is already in the case. That amp looks like a decent mid-fi to start with, but it will still take some DIY to get it up to hi-fi, in my opinion. Start with decent ventilation. Huge error: Air output vents have insufficient airflow unless you also provide an equal or greater size air intake vent at the bottom of the case. I don't see it, unless it happens to be under the PCB (which would be great for cooling off the caps). Also, the hot air output vent on top is too tiny.
Got a drill?
I see dodgy slippy screw connectors that can cause erratic connections, so all those cables need fluxed, tinned and soldered trackside underneath their respective screw connectors (after you've completed any desired component swapping and quality control).
P.S.
I suggest that all of the electrolytic caps need to be swapped for 105c heat tolerant caps. That amp is built like a little oven.
Last edited:
Need all of your valuable suggestion
Hi all
I am very happy and feel privileged to find such exhausive and valuable information about audio by so many experienced members here.
This is my first post here and hence quite unsure where to post my queries.
My main amp is a high power mosfet one. But hearing many good things about TDA2030A, I attempted to build them following the pcb layout in the datasheet. I was so pleasantly surprised to see the result that i wanted to upgrade to TDA2050.
Currently I am using 4 TDA2050 with a 2 way active corssover system with a 3 band equalizer and and a NE5532 based buffered preamp. I had a listening session with my 6 friends who uninamously agreed that they heard a very true to life sound with excellent stereo imaging, surprising minute details that was absent earlier. And we had no prior exposure to any really high end audiophile system to be honest.
The output is more than adequate for my very small listening room. And I play it at a very low volume.
So my question ....
1.Do I really need to upgrade to LM3886 or anything better in my situation, in a near field listening environment? (LM3875 is hard to get in my locality)
2. Is an active 2 way crossover system better than a 3 way system in a small room?
3. I was unable to find OPA2134 yet , so I am using NE5532 for the time being. I want to swap opamps just for a test Could you please recommend some latest commonly available ( better) opamps?
Thank you all,
With warm Regards
Som
Hi all
I am very happy and feel privileged to find such exhausive and valuable information about audio by so many experienced members here.
This is my first post here and hence quite unsure where to post my queries.
My main amp is a high power mosfet one. But hearing many good things about TDA2030A, I attempted to build them following the pcb layout in the datasheet. I was so pleasantly surprised to see the result that i wanted to upgrade to TDA2050.
Currently I am using 4 TDA2050 with a 2 way active corssover system with a 3 band equalizer and and a NE5532 based buffered preamp. I had a listening session with my 6 friends who uninamously agreed that they heard a very true to life sound with excellent stereo imaging, surprising minute details that was absent earlier. And we had no prior exposure to any really high end audiophile system to be honest.
The output is more than adequate for my very small listening room. And I play it at a very low volume.
So my question ....
1.Do I really need to upgrade to LM3886 or anything better in my situation, in a near field listening environment? (LM3875 is hard to get in my locality)
2. Is an active 2 way crossover system better than a 3 way system in a small room?
3. I was unable to find OPA2134 yet , so I am using NE5532 for the time being. I want to swap opamps just for a test
Could you please recommend some latest commonly available ( better) opamps?Thank you all,
With warm Regards
Som
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.