jaudio said:
test,to show u
will show ya the readings i obtained from the measurements of both the amps as soon as i find my note book(of my 2nd year engg.)
freq response,,,,from a signal generater(triangle wave)
i remember that gain for bases was higher for lm with almost(real world conditions) flat response for full band with tda.
no test for 'painted',,listening to both chips one after other ,same song,same drivers,,,,lm sounded lacking some details,woofer content were evidently high(all those who listened said so,many liked heavy bases).
lack of details(compared to tda) is substantiated by the tests,wherein the gain at higher frequencies was subdued.
it seems u r very happy with your lm.thats good.
i m not saying lm is bad chip.its good,but when compared to ST microelectronics' tda, tda emerges as winner.
Just to throw it in: 7294 is a DMOS-based amp, not MOSFET.
Also the application notes state that in the case of requiring more than 50-60 watts of output power the dissipation is not manageable and it needs to be constructed in the high-efficiency circuit, also given on the datasheet.
From the 3886 datasheet, it delivers 68 watts, and 70 watts for the 7294. The real story comes just after that point, where the 7294 allows for higher dissipation and can push instantaneous peaks of upto 180 watts compared to 135 watts for the 3886. The datasheets tell this story.
The datasheets have another story, that is distortion performance. Across the audio band and specially at high frequencies, the 3886 eats the 7294 as far as distortion goes. And this is complicated by adding more parts like a standby, mute etc.
Overall IMO both chips are meant to cater to the same market and reside inside similar appliances. Except for the fact that the 7294 has more external components (and also more configurations such as high-eff due to separated rails) they would be implemented similarly.
As far as failures go, both are statistically able to fail. The chances of failure are similar and there's little point in generalising about a line of chips based on a sample of one (or six), it's just not statistically valid.
Then there's sound quality.
That is one game where no one is right or wrong. To each their own, and I may find the TBA810 the best chip in the whole world as far as SQ goes (it has 5% THD at 1 watt). I don't, but that's different story. I think one should build and test, and be happy with their choice. The original post was asking about the differences between the two. Let's not turn it into a battle for supremacy between the chips, based on subjective opinion.
Also the application notes state that in the case of requiring more than 50-60 watts of output power the dissipation is not manageable and it needs to be constructed in the high-efficiency circuit, also given on the datasheet.
From the 3886 datasheet, it delivers 68 watts, and 70 watts for the 7294. The real story comes just after that point, where the 7294 allows for higher dissipation and can push instantaneous peaks of upto 180 watts compared to 135 watts for the 3886. The datasheets tell this story.
The datasheets have another story, that is distortion performance. Across the audio band and specially at high frequencies, the 3886 eats the 7294 as far as distortion goes. And this is complicated by adding more parts like a standby, mute etc.
Overall IMO both chips are meant to cater to the same market and reside inside similar appliances. Except for the fact that the 7294 has more external components (and also more configurations such as high-eff due to separated rails) they would be implemented similarly.
As far as failures go, both are statistically able to fail. The chances of failure are similar and there's little point in generalising about a line of chips based on a sample of one (or six), it's just not statistically valid.
Then there's sound quality.
That is one game where no one is right or wrong. To each their own, and I may find the TBA810 the best chip in the whole world as far as SQ goes (it has 5% THD at 1 watt). I don't, but that's different story. I think one should build and test, and be happy with their choice. The original post was asking about the differences between the two. Let's not turn it into a battle for supremacy between the chips, based on subjective opinion.
This link below might be of interest as to the sound and capabilities of the TDA7294 chip.
http://www.irdaudio.com/products/mb100.htm
http://www.irdaudio.com/products/mb100.htm
Since you claim you were studying engineering, I'll assume you are correct, but this is what I found:
DMOS is a fabrication process. Double-Diffused Metal Oxide Semiconductor, if you must know the abbreviation (I don't know why HP chose to use a single D). Used to make transistors, switches and diodes as well as FETs and JFETs.
MOS-FET is a specific type of transistor. Metal Oxide Semiconductor Field Effect Transistor. FETs can be made with MOS, CMOS, LDMOS, VDMOS, V-Groove, HEXFET, or IGBT (mixed tech of Bipolar+FET) processes. Therefore you can have a DMOS-FET, LDMOS-FET and so on.
One does not, in my view, equal another. You're the engineer. You tell me.
From the datasheet: To overcome these substantial drawbacks, the use of power MOS devices, which are immune from secondary breakdown is highly desirable. The device described has therefore been developed in a mixed bipolar-MOS high voltage technology called BCD 100.
By that statement I'm guessing they mean this is not a pure MOSFET amp, but only the output stage is made of Power DMOS devices. If it was a MOSFET amp, they would have called it a MOSFET amp.
If I'm not mistaken, this is one type of a hybrid amp (the other being the STK series which mixed integrated and discrete on the same substrate).
Power to you sagar, I'm sure you already know all this and are right on the money about this stuff.
DMOS is a fabrication process. Double-Diffused Metal Oxide Semiconductor, if you must know the abbreviation (I don't know why HP chose to use a single D). Used to make transistors, switches and diodes as well as FETs and JFETs.
MOS-FET is a specific type of transistor. Metal Oxide Semiconductor Field Effect Transistor. FETs can be made with MOS, CMOS, LDMOS, VDMOS, V-Groove, HEXFET, or IGBT (mixed tech of Bipolar+FET) processes. Therefore you can have a DMOS-FET, LDMOS-FET and so on.
One does not, in my view, equal another. You're the engineer. You tell me.
From the datasheet: To overcome these substantial drawbacks, the use of power MOS devices, which are immune from secondary breakdown is highly desirable. The device described has therefore been developed in a mixed bipolar-MOS high voltage technology called BCD 100.
By that statement I'm guessing they mean this is not a pure MOSFET amp, but only the output stage is made of Power DMOS devices. If it was a MOSFET amp, they would have called it a MOSFET amp.
If I'm not mistaken, this is one type of a hybrid amp (the other being the STK series which mixed integrated and discrete on the same substrate).
Power to you sagar, I'm sure you already know all this and are right on the money about this stuff.
Nay9 said:
Norh used to make an amp called the LeAMP which had the chip. Peter Daniel had also tested the 7294 with decent results, though he eventually went for the 3875 (not 3886) in later implementations.
So yes, the chip has promise if implemented well.
Whether it can beat the 3886 or not (as asked by the original poster, who's kind of been forgotten in all of this) depends on too many factors...
no claim,i m in final year(elect.) now.
but what is the end result of the process,its a FET ,metal oxide semiconductor FET.MOSFET.
these different processes determine how the conducting channel will form in the MOSFET.this allows for MOSFETs being designed for requisite applications.
DMOS process in tda case results in a high power and high voltage mosfet.
for tda input stages r bjt based,the main or the o/p stage is MOSFET made using DMOS process.(to enable high voltage and wattage along with inherrent benifits that MOSFET offers over bjt)
see block diagram of tda i post here
Attachments
jaudio said:
In the end it's a matter of taste.
But to take that conclusion one must build both amps according to what's best for each chip.
These chips are quite different and requiere different implementations.
Jumping to conclusions like 'this one is better' is precipitated unless one has some practice with each of the chips and has lost some time trying to get the most out of them.
Op-amps are not 'universal', even the best op-amp you know can produce poor results on a certain circuit.
Change the circuit and optimize it for the op-amp in question and voilá.
Bipolars are different from fets and require different design.
Then there's noise, bandwidth, speed, dc-offset considerations that make have to change things accordingly.
For instance, the implementation for an OPA541/8/9 has to be different from the LM chips, or they will perform very bad.
Then there's the PSU, which is very important too.
Conclusion: I would not say a TDA is better if I didn't know it very well, tested several implementations and made several different amps.
It is a good chip alright, Linn uses the TDA7293 and has used other TDA chips in the past.
Those amps are not bad at all, but still not up to my LM3886.
I never bothered to try the TDA7294/3, so in practice I can't give an answer.
I just made some remarks about the specs, and also about the idea that the LM3886 has overblown, unnatural bass.
Not true!
sagarverma said:
That is your conclusion.
Not mine, and not everyone's.
You simply couldn't put the LM3886 sounding neutral, detailed, 'airy', with TIGHT bass, as it can provide.
So, why should we (or I) take your oppinion as a matter of fact?
It is your oppinion, based on the amps you built with these chips.
Don't generalize, because a chip can sound very different, depending on the implementation.
Ignoring this is ignorance.
I built a small amp using Original TDA7294, it was fairly good but did not offer the vocal quality of the LM's, I also tried the 7293 with the bootstrap cap, this was also not bad but behind the LM IMO
Still these chips are nice enough IF you can buy the originals
Theres a lot of cheap TDA7293/94 copys floating about now, they are not as reliable as the original, I've seen a few with a hole blown out in the middle
Still these chips are nice enough IF you can buy the originals
Theres a lot of cheap TDA7293/94 copys floating about now, they are not as reliable as the original, I've seen a few with a hole blown out in the middle
carlosfm said:
That is your conclusion.
Not mine, and not everyone's.
You simply couldn't put the LM3886 sounding neutral, detailed, 'airy', with TIGHT bass, as it can provide.
So, why should we (or I) take your oppinion as a matter of fact?
It is your oppinion, based on the amps you built with these chips.
Don't generalize, because a chip can sound very different, depending on the implementation.
Ignoring this is ignorance.
plz check the posts properly b4 posting anything.
i said MY opinion,NOT generalised.if u couldnt get that right,
Guys:
May i suggest one thing, experience with chips may vary from person to person, for sagar TDA729x sounds better than LM3886, and carlos himself has always been an advocate for OPA627 over AD8610, whereas opinion of many people with the same chips (including me) is opposite, so if sagar thinks that TDA729x sounds better than LM3886 for him then is there any point in proving or trying to convince hime that he is not right.
One's personal experince overweighs the suggestion or experince of anyone else.
May i suggest one thing, experience with chips may vary from person to person, for sagar TDA729x sounds better than LM3886, and carlos himself has always been an advocate for OPA627 over AD8610, whereas opinion of many people with the same chips (including me) is opposite, so if sagar thinks that TDA729x sounds better than LM3886 for him then is there any point in proving or trying to convince hime that he is not right.
One's personal experince overweighs the suggestion or experince of anyone else.
Sagarverma,
In my opinion, some of your posts are rather 'blunt'; perhaps, you could express yourself a little more politely and with finese. This is an open forum, everyone's opinion is welcome and it is better not to get hung up on 'who is right' because the quest here atleast technically is to better the 'what is right' established premise. Hence, in the bid to help one and all to greater audio building and listening experiences, it is only mandatory to keep the spirit of this forum.
Secondly, it would be right for you to post the schematics and layouts of the respective circuits you tried out for others to evaluate your conclusions more objectively.
In my opinion, some of your posts are rather 'blunt'; perhaps, you could express yourself a little more politely and with finese. This is an open forum, everyone's opinion is welcome and it is better not to get hung up on 'who is right' because the quest here atleast technically is to better the 'what is right' established premise. Hence, in the bid to help one and all to greater audio building and listening experiences, it is only mandatory to keep the spirit of this forum.
Secondly, it would be right for you to post the schematics and layouts of the respective circuits you tried out for others to evaluate your conclusions more objectively.
Samuel Jayaraj said:
plz mention my blunt posts.i sincerely had always tried to be polite to keep the spirit of this forum up.
posting of schematics not req. in any post here till now(plz read all post till now properly).i have posted images etc wherever applicable.no use of posting unnecessarily when its not req. or others dont demand.
carlosfm said:
Has already beaten??!
If it 'forgives errors' and works better for you, by all means, use it.
samuel,what would u call this?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.