• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Not 15A: 150A transistor req'd for 75-150W class AB amplifires with low THD+n over 12W? OP+some of post 9 req'd reading to understand subject first

Status
Not open for further replies.

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
Update: Still looking for default distortion profile when switching from lower powered transistor to higher power (with all other parameters equal).

Generally it's been brought to my attention that there are many variables that affect distortion, but the precise effect having a more powerful transistor in place (capable of higher max current) has on distortion - I'm still looking for

Because I changed the title, I need a preamble to explain a bit. Look at the chart below in OP.. you'll see that all 4 Denon AVRs ranging 105-150WPC and 9-11 AMPLIFIED channels, all begin distorting more starting at 12W. In my reading up to this point I feel comfortable stating all 4 receivers use 15A transistors. OP and post 9 explain the significance of this - I believe a big part of the issue this thread aims to tackle is that transistors require quite a bit of feedback to remain linear once pushed past 8-10% of their maximum rated current (for example, a 15A transistor at 50C behaves well (requires minimal feedback for output into resistive load to behave - up to 1.2-1.5A).
If you knew nothing of what I just said, you might think that if you had two very similarly designed AVRs, one rated for 105W RMS, and the other rated for 150W RMS, and you saw the 105W AVR's THD+n began increasing from 12W, you might hazard a guess that the 150W receiver might start distorting at 18W, because they're a similar designj, but the 150W receiver is capable of about 1.5X more power than the 105W. You'd also be wrong, because both receivers THD+n begin increasing at 12W.
Another interesting fact is that this distortion begins at 12W whether the load is 8 ohms or 4 ohms...
Yeah, that one's kind of a bombshell lol.
This evidence suggests the reason for the observed increased distortion after 12W is the transistor operating outside of its 8-10% zone previously mentioned. I have a class A amplifier I discuss later in OP/p9 which uses 10 transistors in parallel to accomplish operating with minimal feedback.
This thread is to brainstorm ways to reduce the amount of feedback required in push/pull designs. Obviously class A is easier to stick a bunch of transistors in parallel and call it a day.



Are we using greatly underpowered transistors in our class AB amplifiers? (the push-pull design biased for minimal crossover distortion)

I think we might be!

The reason is... actually, there are many. But to start, I'll begin with the thing most at the forefront of my thoughts - I have a Kinergetics KBA-280 - a stereo class A amplifier rated for 140W RMS into 8 ohms. It's a behemoth - it weighs more than any other amplifier I've lifted, and the transformer is a toroid. Inside there are... I want to say... 10 output transistors per channel, and they're not small. I forget what they are right now, but their package is like 1.8cm x 3.5cm x 0.4cm. They're MASSIVE transistors, and the amount of them in there? Is MASSIVE!
Stereophile reviewed it, and during their review, worked it HARD. How hard?
Single ohm hard.

Why? I don't know, probably had something to do with the weight of it and they're connoisseurs. Next is watts into 8, 4 and 2 ohms.
140, 270, 530, and it gets even better when the final power rating is taken while driving a single ohm...
A THOUSAND WATTS PER CHANNEL. But only one at a time, because the main fuse.
Right now we're talking about something else though: this thing's transistors. We can marvel later

For the sake of discussion, let's say behemoth's transistors are rated for 15A max. They're probably 15 amps (if they're not, they're more)

15A x 10 = 150

150 amps

That's a TONNE of amps! More than the service to my last place! (which I believe was 120A).

150A and an 8 ohm load? That's 1200 volts.

1200 x 150 = 180,000W

I looked at the datasheet and found out: where the transistor is most linear is in the first 7-10% of its "maximum" rating. Past that, it's pretty nonlinear. How does this nonlinearity increase? Nonlinearly! Funny, isn't it, that the increase in nonlinearity of a transistor is nonlinear itself! Lol. It kinda looks like y=a^x ("y" equals "a" to the power of "x").
Fortunately this unfortunate part of transistor existence doesn't affect its operation most of the time, because in addition to the transistor's input, feedback (how far off the transistor is from its intended path) is also applied, simultaneously. This ensures that the output becomes what's desired, and not just the transistor's nonlinear representation of its input. How is this feedback collected and how is it applied to linearize? It varies a lot, and can get complicated. Which is why, if you can, you minimize the amount of feedback required :)
And in the end there's less parts, too.

So moving on... let's use 8% of 150A

12A

with an 8 ohm load 1152 watts, and 96 volts

I think the transistors were 200V, which in class A gives a maximum of 70VAC. Since you can't hit the rails, call it 65VAC

528 watts, and linear behaviour to over 2x that amount of current.


From what I gather, these outputs would need minimal feedback because the transistors are operating well within their optimal operating range (that small fraction of their maximum rating).



There are 10 transistors being used to do the work that just one transistor can do.

The only real caveat to performance from running all 10 transistors in parallel in the way I've outlined (class A) is the noise floor would increase by 10dB. But even half decent transistors are pretty quiet, and 10dB is nothing compared to the benefits from running in linear territory.



Now the conversation is shifting though- background is done. The graphs below, though - they need to be considered as the first step to the next direction. Four of these charts are power vs. distortion of recent Denon home theater receivers. The fifth more bland looking one? Still from this site, just someone else measuring a Yamaha that has the same type of amplifier (AB), but it's rated for 60 watts RMS x2 instead of 105/125/140/150 watts (Denon X3700/4700/6700/8500H). Yamaha got a slot so that all the amps weren't Denons, and there was one significantly different. I believe all of the Denons were made with 15A transistors.
Aside that needs to be said that I wish I didn't because it's annoying: SO, the fourth Denon (X8500H) is represented by the 8 ohm chart instead of the 4 ohm chart like all the others, because the X8500H's 4 ohm chart is ugly as sin. One channel was very divergent, and ignoring it and putting a line through the other is stupid when the better looking 8 ohm chart is there to use. It shows the same result as the 4 ohm chart and does it lookin' good :cool:
'

1674840353239.png



3800 4800 6700 8500

Above you can see that the X3700H, X4700H, X6700H, and X8500H, are all capable of different maximum continuous stereo power outputs.
And you can see that they all four beasts are capable of the same amount of power before THD+N stops improving with increasing power.

WHY IS THIS?

8% of 15A is 1.2A

1.2A and 8 ohms is 11.52 watts.

The charts above, the vertical green lines I've marked between 10 and 20 watts, more accurately are between 12.2 and 14 watts.

Before accurately calling the change in a trend, a marked difference needs to happen. Those lines at 12.2 and 14 are after a noticeable change - a clear trend. If you look to precisely where the green line drawn from ~100mW to ~2000mW diverges.......

11.5W


11.5 watts x 2 channels is well below the maximum current able to be supplied by the power supply

11.5 watts x 2 channels into 8 ohms or 4 ohms requires much less than the amount of voltage supplied to the transistors

The new receivers for 2023: the X3800H, X4800H, X6700H, and X8500HA ALL have 15 amp transistors. I can't say for sure that the models above also have 15 amp transistors, but I know the X3700H does (I have one, wooo! partyy!), and I'm fairly sure the X4700H does. So if the X8500HA does too, I think it'd be stupid to think the 6700 didn't!


I think the reason distortion goes up at about 11.5 watts is because more feedback is needed.


Does anyone who knows more than me think what I described is plausible?
If so, what can we do moving forward designing amplifiers?

Putting a bunch in parallel works for class A amplifiers. For low distortion, class AB (working in B, push-pull) need to be biased to minimize crossover distortion as much as possible or none of the rest matters. If using multiple transistors in parallel for push pull, each transistor has a different voltage that it turns on. This voltage also changes with temperature. Question: if two turn on at the same voltage at 20C, will they both turn on at the same voltage at 50C? 90C?
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
For the sake of discussion, let's say behemoth's transistors are rated for 15A max. They're probably 15 amps (if they're not, they're more)

15A x 10 = 150

150 amps

That's a TONNE of amps! More than the service to my last place! (which I believe was 120A).

150A and an 8 ohm load? That's 1200 volts.

1200 x 150 = 180,000W

One should realize that it is not just the Amps and Volts in a data sheet that matter.
The SafeOperatingArea is what matters so the power rating.
And even this is with optimum cooling.

A 15A 100V rated transistor is not capable of delivering 1500W.
It can carry 15A but with hardly any voltage over it and can withstand 100V but not while conducting a high current.
Such a transistor could well be 90W and at 30V across it that means 3A. With excellent cooling that is.
An amp that can provide 1kW in 1ohm, 500W in 2ohm, 250W in 4ohm and 125W in 8ohm can provide 31Vrms which requires +/- 48V rails.
At 1ohm there will be 31A(rms) drawn with peaks 44A.
When this is in class-A (up to a certain load) this means it will continuously draw somewhere around 1.5kW per channel (assuming 15A idle) in heat alone.
With each transistor handling say 50W you will need 30 transistors (15 NPN and 15PNP) when they also need to deliver power in 1ohm occasionally.

Question: if two turn on at the same voltage at 20C, will they both turn on at the same voltage at 50C? 90C?

The transistors do not turn on, they will start to conduct more current when the output voltage increases.
When multiple output devices are in parallel they also have emitter resistors which ensure the current is always distributed evenly across the output devices.
Distortion is reduced by overall feedback and proper bias current (when no signal is present).

Only in switching output stages (class D and T and such) the transistors switch on fully and off.
There you can have close to 0 volt across the transistor at max current and when no current is delivered the max. voltage across it.
That's why they are so efficient and do not get hot as little power is demanded.
So a LOT of power, small footprint (heatsinks) and not many output devices needed.

Of course, as transistors need time to conduct fully and time to stop conducting there is still more heat dissipated than one might think and this depends on the speed of the transistor, the drawn current and the switching frequency.
 
Last edited:
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
One should realize that it is not just the Amps and Volts in a data sheet that matter.
The SafeOperatingArea is what matters so the power rating.
And even this is with optimum cooling.

A 15A 100V rated transistor is not capable of delivering 1500W.
It can carry 15A but with hardly any voltage over it and can withstand 100V but not while conducting a high current.
If you read my post I think you can infer I realise this.
Unless you're reminding potential brainstormers?
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
One should realize that it is not just the Amps and Volts in a data sheet that matter.
The SafeOperatingArea is what matters so the power rating.
And even this is with optimum cooling.

A 15A 100V rated transistor is not capable of delivering 1500W.
It can carry 15A but with hardly any voltage over it and can withstand 100V but not while conducting a high current.
Such a transistor could well be 90W and at 30V across it that means 3A. With excellent cooling that is.
An amp that can provide 1kW in 1ohm, 500W in 2ohm, 250W in 4ohm and 125W in 8ohm can provide 31Vrms which requires +/- 48V rails.
At 1ohm there will be 31A(rms) drawn with peaks 44A.



The transistors do not turn on, they will start to conduct more current when the output voltage increases.
When multiple output devices are in parallel they also have emitter resistors which ensure the current is always distributed evenly across the output devices.
Distortion is reduced by overall feedback and proper bias current (when no signal is present).

Only in switching output stages (class D and T and such) the transistors switch on fully and off.
There you can have close to 0 volt across the transistor at max current and when no current is delivered the max. voltage across it.
That's why they are so efficient and do not get hot as little power is demanded.
So a LOT of power, small footprint (heatsinks) and not many output devices needed.

Of course, as transistors need time to conduct fully and time to stop conducting there is still more heat dissipated than one might think and this depends on the speed of the transistor, the drawn current and the switching frequency.

I appreciate the help, but I have to say you're missing the point.
I've worded things so that most people should be able to understand the problem. Eg. I know the transistors don't turn on, but they do effectively start working..
Eg2. I know how many volts are needed for 140/280 watts into 8 ohms etc. I was explaining the maximum amount of power into 8 ohms that 200V peak to peak can make - essentially, no matter how many transistors were put in parallel for that class A amp, the limit (528W)can't be breached.

I know that the maximum rating of a transistor isn't what it's supposed to be used continuous. It's maximum. A 15A transistor is fine making 15A for 20ms, 0A for 980ms, forever. Or it's fine making music into a resistive load where peak current reaches but doesn't exceed 15A, while the heat it generates from the average power of the music is adequately dissipated.

These aren't the things being discussed though, or other types of amplifier. Just Class AB, while in B mode, push-pull, and biased for minimum crossover distortion ("on" at 0V). The idea is to theorize ways to minimize THD+N of push-pull transistor amplifiers, taking into account that the increase observed in existing amplifiers is caused by requiring all that feedback because the transistors are operating in their non-linear range

I know none of their operation is perfectly linear, and that some correction will be required. Especially in push-pull.

Goal: Modify the basic AB amplifier design so that minimal feedback will be required during operation (especially at high volume. obviously)

I already stated a couple detriments to using two transistors in parallel to increase current, at the end of the OP. A solution to that is also progress.

Edit: Or telling me that I'm wrong and the reason for the increase in distortion isn't because the transistors are being operated so far out of their linear range and are requiring feedback. But if you tell me that you have to be right, obviously. Right now I'm quite sure I'm right about that
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
What are you trying to solve ?
It is perfectly possible to design a high power amp with low noise and distortion.
The amps in AVR's do not have this goal nor is it needed.
They are just limited by design, functionality and price point.
They don't need to provide continuous power, they just have to play music/movies and drive a bunch of speakers.
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,191
Location
Northern Virginia, USA
What are you trying to solve ?
It is perfectly possible to design a high power amp with low noise and distortion.
The amps in AVR's do not have this goal nor is it needed.
They are just limited by design, functionality and price point.
Right? Since when was an AVR a state-of-the-art amplifier? I mean they're great value...but...
 

dorakeg

Senior Member
Joined
Jul 20, 2022
Messages
326
Likes
187
What are you trying to solve ?
It is perfectly possible to design a high power amp with low noise and distortion.
The amps in AVR's do not have this goal nor is it needed.
They are just limited by design, functionality and price point.
They don't need to provide continuous power, they just have to play music/movies and drive a bunch of speakers.

Yes agreed. AVRs are never designed for huge power. For those who desire power, they will go for avp/power amp combo instead.
 

kongwee

Major Contributor
Joined
Jan 22, 2022
Messages
1,024
Likes
276
Your power consumption rated at the back of your AVR will never tally with the power of rated channel output.
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
I referenced those AVRs because the amplifiers in them are first rate, and, especially for two channel playback, the power supplies in them are about 3 times the size they would be if the device had been designed for only stereo playback at its rated power output. And these aren't wimpy multichannel receivers, either - you know the kind I'm talking about that are - claiming 100W + 100W at the fronts and 60W to your centre, and you get 20W + 20W for the rears! No, no more of that early 90s garbage they only got away with because the surround channels were decoded from stereo. All these Denons, they amplify 9 or 11 channels, and each one of those channels uses the same amp section. It's hard for some people to come to grips with the fact that the fronts are the same as (perform the same as) the rears, are the same as the front heights, are the same as the centre, are the same as the surrounds, are the same as the rear heights, are the same as those extra two channels you configured to use in the bi-amp configuration to power your fronts because you built them yourself and and like to make your amplifiers work, so you designed them with two impedance dips below two ohms - one smack in the middle of the punchy bass range: 90Hz is down 1.8 ohms. The other one is worse but in the subsonic range so there's not much energy down at 23Hz - it doesn't matter.

OK, I'm done being funny. To the point:

Looking at measurements of these receivers, it appears the basic foundation for the amplifier section is shared across Denon AVR models - with the more premium designs using higher quality passive components, higher powered active components, higher voltage transformer secondary outputs, and more of and closer attention paid to the further refinements necessary to implement the amplifiers properly in an AVR. I'm not guessing at the quality and power and voltage - Denon releases info about the design of their stuff from time to time. Recently I came across (I think on a French site) proof of exactly this.

Consider the flagship model (X8500HA) - it is larger and heavier than the other receivers (X3700H, X4700H), which are mid-upper tier models. How much bigger is it? To the eye it appears about 75% bigger. Why? Mostly because it has 20% more channels that are 50% more powerful. When you take this into account, and that the flagship has many features and functions the lesser models don't (including more inputs), there likely isn't a lot (or any) extra room/space to be creative with how they choose to spiff up the performance of the amplifier section. An interesting suggestion: individual 6,800uf capacitors for each channel board in the X8500HA to ensure exceptional performance re: transients, even from the channels furthest from the 33,000 bulks. "Are ya stoopid?" came out of the project manager's mouth at the idiot who suggested it. "Where is there space for ELEVEN 80V low-ESR audio-grade capacitors in this chassis?! Are you THICK?" Do what you did last time: LCR closest, the others following in their order of likelihood receive dynamic signal."
And consider: this capacitor is just a single component. Three or more are needed for most meaningful modifications - and when you multiply that by the flagship and 2nd place's 11 channels, you get thirty-three (33) more components for an amp section that's definitely different, but depending who you're asking or what the situation is, the mod may have a negative overall effect on sound quality. Also with these modifications there is the risk of unintended [side] effects, plus there are more parts to break.

OK, so:

Class AB amplifiers, regardless of manufacturer, when operating push-pull configuration (not in class A mode for 1-2 watts and less...)at a fraction of a watt), exhibit increasing distortion as the output transistor's current reaches 8-10% of its maximum rating. More and more after this threshold is past, transistors put their own spin on their input - unless they're corrected. How are they corrected? With feedback. Yes feedback is always used, but the extent to which it's necessary pre 8-10% is next to nil compared to 20%. And 50%? Look out because without feedback, output's so different than desired, it's close to unrecognizable.

Think of having two people on either side of one person in the centre. The job of those wo people is to keep centre person on track. When walking, centre person keeps to where he's supposed to go. It's any faster than walking that he starts to drift. Into left person, back to track. Into right person, back to track. Left/right person guide centre person back on the track, but centre reaches left left person. Left person isn't infinitely strong or infinitely fast, and he doesn't get instructions of the angle to shove Mr. Centre at until it's known the angle at which Mr. Centre left his lane.

-At any pace faster than walking, if Mr. Centre loses his centre he will not end up where he's told to go. And nobody's perfect - even though Mr. Centre is an accomplished walker, all it takes is just needs to slip a little too far one way once, and the track, what track? He's completely and for good headed in the wrong direction until left/right give direction.

-With Mr. Centre kept on track by left and right, Mr. Centre always ends up on the path the entire time and gets to the end. But left and right are even more fallible than their physical limitations - they're completely dependent on their directions. They get directions from someone like Mr. Centre does, but they're much better at following them - almost perfect, in fact. Why doesn't Mr. Perfect just do Mr. Centre's job? Well, Mr. Centre is specialized, and the ground beneath his feet is in 3D, while his instructions are in 2D. Mr. Centre balances and navigates and endures the ups and downs, side to sides, the inclement weather, and the rest of it. When Mr. Centre slips, Mr. Perfect hears about it and gives out instructions to left or right to execute for the course correction Mr. Centre needs to be right again.

Why does Mr. Centre slip?
Because his feet don't grip the ground strong enough for speeds faster than walking. He's not stupid like you might have thought, he's just too weak to go faster than walking without constant help. The faster he goes, the more often and more violent the bumps will be. He can't travel on his own because the 3D terrain is mostly benign, but occasionally though quite consistently there are anomalies or aberrations in his path which he is not able to just continue through without a course change which requires left or right person to set him right again.

What can be done to make Mr. Centre not slip?
We clone him 10 times without a brain, and then we make those 10 beings occupy the same physical space. This way he's 10-12 times stronger and doesn't have to fight for control with NPC versions of himself. The way this is done is very complex, but just think of putting some electronic devices in parallel, like a transistor:
Because each transistor is an exact copy, they have the exact same characteristics, so all 10 transistors are arranged on a circular heatsink so that they operate at the exact same temperature. This way they are able to function as a single transistor that's 10dB noisier but has the amazing property of being capable of 10x the current of one

That concludes the story of Mr. Centre lol. I hope you get what I was getting at now. I had an interesting time trying to make an analogy that didn't quite work, work.
Well enough lol.


Moving onwards:
Denon's amplifier sections lately are great. I have good ears and am a very experienced listener. I've also listened to a variety of systems and amplifiers. Between 6 and 3 years ago I was looking for an amplifier I'd be happy with that I could also afford. I tried many things... Things that got amazing reviews I found sounded narrow and closed in. Things that are supposedly similar (Arcam A18 and A19) do not sound similar at all. A friend of mine had an A18, and I heard it and was like "WOW!, the detail! The width of the soundstage! The frequency extension to infinity! The lack of listening fatigue! It's.... perfect!!!" I wanted one so badly, but they stopped making them! I went to the local hi-fi shop and was told they make the A19 now- it sounds even better!.
Liars. Liars liars liars. $1500CDN later and the detail was missing, the top end extension - not there, soundstage width: it was there, but underwhelming when paired with the listening fatigue after just 20-30 minutes of moderate listening.

Amplifiers are hard to get to sound great. You can study theory and draw up some plans for an amplifier, assemble it, connect it, power it on, and listen to music through it... but it's not going to sound as nice as the like the Arcam A18.

I sold my A19 for $375 and found an Arcam A38 to buy. The A18, A28, and A38 are a series. I was still a bit "new" at the time (new to knowing), so I thought the A38 would be the 150 watt version of the 50 watt A18. Boy was I wrong... The A38 was definitely better than my A19 - more detail, better bass, better extension..., and it did seem to be "voiced" like the A18, but it.. was.. no.. A18!

Because a 150W amp is a vastly different beast than a 50W amp, literally all integral parts were different between them. Because the A38's sound signature was similar while the overall sound quality was worse, I had the thought that maybe some of the parts that the engineers planned to use were vetoed by the money people.
"You don't need a shielded cap here!"
"Schottky rectifier? Why spend the extra $5.32? You've got those expensive massive Nichicon..."
"A resistor is a resistor! Wire wound in the signal path is fineeee. We can save 4 cents per unit for the 3 resistors. That's $1600!!"
-3 days before production run "Sorry Jan, .22mH were only available in 20ga. Since .22mH and 22ga was specified, I ordered the next closest size in 22ga: .15mH. Cheaper, too!
"Bulk capacitors will not be Nichicon 22,000uf, they will be DIMSUNSU 15,300uf. ESR will be 2.5x higher"
"Remember when you were experimenting with trains of ceramic caps of various sizes to eliminate 99.7% of remaining SMPS noise at amp output? Remember when you ordered some custom surface mount packages made of the sizes you decided on - all stacked right beside each other essentially in one device? Well, that felt pretty wasteful to me this afternoon - I went into the studio and A/B'd the unit with and without your filter and there was no difference! I cancelled the order... You're lucky I don't write you up for cavalier attitude toward spending company money."
reply "Uh, boss... 99% of the time the studio is powered by the thing we have that regenerates AC power so that we're not using what comes from the power company directly? This way we can isolate noise caused externally from noise caused internally. Once the product is quiet, we make sure it remains quiet when hooked up to noisy power. You literally just thwarted that last step"
boss's reply: "Too late now! Good thing I'm the boss and brand recognition is 90% more important than how our stuff sounds now that we're established!


I'm getting distracted lol.

Essentially $20K amps don't have power supplies as good as the ones the Denon amps have because the Denon amps have 5x the number of channels.

The increasing distortion with volume past ~12 watts cannot be from the power supply voltage sagging or becoming ripply/pulsed DC.


I think the reason that MOST PUSH-PULL AMPLIFIERS EXHIBIT what's shown in the graphs in my OP (THD+N/distortion increasing from about 12 watts-14 watts) is at that point the transistors beginning to operate above 8-10% of their maximum rating, below which is where operation is quite linear and minimal feedback is required.

I think that, like my class A amplifier which uses multiple transistors in parallel to always operate below 8-10% of their maximum rating, ensuring minimal feedback is required for accurate amplification.

I think that we should try to tackle the problem by thinking about it - people who know more than I, people who design amplifiers:

How can this problem be fixed?
Is it just that the transistors we have are too small and all they have to do is develop a bigger chunk of semiconductor when manufacturing transistors so that their theoretical maximum current is 150A? And even though there's no way the chip could EVER dissipate 150A, the special transistor could be used solo for the "push" or "pull" of an approx 100-200W into 8 ohm amplifier? Basically treated like a single 15A transistor would be and often is - typically operated in an amplifier to an average of 2A with transients peaking at 4-5A (8 ohm)

OR

would smaller transistors have to be rigged up together with "hacks" and "fixes" to get them to behave as one...
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
Your power consumption rated at the back of your AVR will never tally with the power of rated channel output.
I honestly have no idea what I said that made you think I need to be told that my AVR's measured power draw from the wall doesn't match its rated output.

When a post is more than a paragraph long, please try to quote what you're replying to if it isn't immediately evident to an 8 year old. [not to dumb it down (too much lol), to remove ambiguity]

If you don't have time or patience to quote because you'd have to snip above and below the section in question, start your reply with restating what it was that I said that you're replying to. Like this:

"about how you said ... and ...." --- When I read this part I'll know what it is you're talking about. Then, your next sentence:
"well it seems .... and ... | if so and ... and... then maybe try..." This way I'll know a bit about what you're thinking and why, and in my next post I'll be able to say something potentially useful, instead of... well... this post. And finally, if you needed clarification on anything in my post you'd replied to part of, bring it up here
"oh, also ... when you said ... in the last paragraph, were you still talking about ... or had you moved on to ... | I'm asking because if you were still talking about ... it doesn't make sense, but if you'd moved on to ... well, that seems really unlikely too" And then as much discussion that needs to occur, can occur. And it won't take a dozen back and forths
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
How can this problem be fixed?

This had been 'fixed' decades ago and it is known how to make good measuring amplifiers.

You still don't seem to understand SOA of transistors and allowed heat dissipation and the impact that has on SOA.
The 'thing' you are looking at is circuit, layout and component dependent not inherent to class-AB designs and not inherent to the number of OP devices in parallel.
Your class-A has that many because the heat that is generated needs to be spread about multiple devices so SOA is never exceeded.
When the amp in question also has to deliver into 1ohm (almost a short) even more output devices are needed to ensure SOA of each device is not reached.

AVR's are a compromise not a high performance high-end device built for impressive numbers.
They are feature packed and this means compromises.
There is nothing to 'fix'. That what you think needs fixing is already known and higher performance is not a problem. Just not in relatively cheap AVR's
The circuit is designed this way with constraints in price, size, component choices. That determines the distortion at higher power levels.
Sure... a part of it is determined by the output devices and amount of overall feedback, but it looks like you think the solution is to add more output devices in parallel which will not help.

A better design is the key. AVRs, though far from measuring and performing exemplary are good enough for the public that uses them.
They won't satisfy folks that want top notch performance nor are they intended for that specific type of customer.

Paralleling OP devices is NOT going to make AVR's perform much 'better', maybe 1dB or so gain, not worth the parts. These amps usually are 20dB away from SOTA.
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
This had been 'fixed' decades ago and it is known how to make good measuring amplifiers.

You still don't seem to understand SOA of transistors and allowed heat dissipation and the impact that has on SOA.
The 'thing' you are looking at is circuit, layout and component dependent not inherent to class-AB designs and not inherent to the number of OP devices in parallel.
Your class-A has that many because the heat that is generated needs to be spread about multiple devices so SOA is never exceeded.
When the amp in question also has to deliver into 1ohm (almost a short) even more output devices are needed to ensure SOA of each device is not reached.

AVR's are a compromise not a high performance high-end device built for impressive numbers.
They are feature packed and this means compromises.
There is nothing to 'fix'. That what you think needs fixing is already known and higher performance is not a problem. Just not in relatively cheap AVR's
The circuit is designed this way with constraints in price, size, component choices. That determines the distortion at higher power levels.
Sure... a part of it is determined by the output devices and amount of overall feedback, but it looks like you think the solution is to add more output devices in parallel which will not help.

A better design is the key. AVRs, though far from measuring and performing exemplary are good enough for the public that uses them.
They won't satisfy folks that want top notch performance nor are they intended for that specific type of customer.

Paralleling OP devices is NOT going to make AVR's perform much 'better', maybe 1dB or so gain, not worth the parts. These amps usually are 20dB away from SOTA.

Thank you for rephrasing and condensing my posts.
I appreciate less the language - fewer people will be able to engage with and postulate. We need solutions!

I didn't realise that the problem had been solved entirely already... so many amplifiers are only really good to 12 watts... some of them double that with the same power ratings. I assume those use two transistors in parallel and are a bit more tizzy from crossover distortion.

Anyway... I presume these people who have fixed this pervasive problem have solved this feedback issue I'm to brainstorm.
But I'm still skeptical with speakers being inductive loads and all... it's gotta be sloppy. Maybe AI + velocity sensors on woofers and current/voltage, (and why not temperature and humidity) monitoring will allow for a pre-feedback'd signal to be generated

I think the easiest way would be to drive a sealed speaker which is critically damped, maybe Q 0.9
Amplifier design would use minimal feedback + powerful single transistor

Do you know enough to theorize if a 150A rated (electrically) transistor being made (a big f-er) to be used up to 12A (not 150A for thermal dissipation reasons), would be viable?

Is there anything wrong with just scaling up the size of a transistor so that a 15A transistor becomes 10x the volume: voila! 150A transistor!!!
I understand past a certain size things don't scale the same way. But this transistor isn't for passing dozens of amps, no. Much smaller.

It'll be used in something like the Denon X3700H. The maximum average current possible will be ~3.62A. Clean (non clipped) music playback would probably have that average current drop to 2.2-2.4A, with transient peaks of ~5A.

Just like the X3700H now, but better!!

Just the transistor would be:
10x larger (physically)
rated for 10x current (150A) but not be able to dissipate it thermally - this would allow function like a 15A transistor, but instead of requiring massive amounts of feedback at just 8-10% its current rating, its entire thermal range it will operate requiring minimal feedback.

And "cheap" amplifiers will, with minimal components have their THD+n decreasing until 120W instead of 12W, covering their entire operating spectrum, making affordable amplifiers just as good as the rest of 'em

We just need bigger transistors, right?
Riiiiight?
...




edit: I don't know if you got what I was saying about how the Denon amplifiers (for sure at least in the X3700H), sound excellent.
You might've missed in my novel that I've listened to systems that cost a lot. Like I said there, too: I'm a practiced listener. Sure THD+n of even the X8500HA grows after 12w, but most listening is done below that. Just pretend they're 12W amps that clip veryverygracefully.



EDIT2:
I think it would be amazing if instead of $4 transistors, they cost $40 and were 10x bigger with the properties I outlined. The circuitry Denon uses in the x3700H could be the default, with these 150A (electrical) 15A (thermal) transistors {10x bigger... maybe 25A thermal :cool:} and everyone could have a amazing amplifiers much better than their speakers. They'd be pretty cheap to make, very simple to repair, just as simple to upgrade. By upgrade I just mean new board, thicker traces, lower resistance caps, higher current transformer, maybe higher rail voltage.

I didn't apply what class G/H could do for a push/pull with one of these transistors!

At the very worst, peak power for short bursts (like explosions) could be increased from the usual 75-150, maybe 220W with active cooling, to 220W passive, 300W active
 
Last edited:

kongwee

Major Contributor
Joined
Jan 22, 2022
Messages
1,024
Likes
276
I honestly have no idea what I said that made you think I need to be told that my AVR's measured power draw from the wall doesn't match its rated output.

When a post is more than a paragraph long, please try to quote what you're replying to if it isn't immediately evident to an 8 year old. [not to dumb it down (too much lol), to remove ambiguity]

If you don't have time or patience to quote because you'd have to snip above and below the section in question, start your reply with restating what it was that I said that you're replying to. Like this:

"about how you said ... and ...." --- When I read this part I'll know what it is you're talking about. Then, your next sentence:
"well it seems .... and ... | if so and ... and... then maybe try..." This way I'll know a bit about what you're thinking and why, and in my next post I'll be able to say something potentially useful, instead of... well... this post. And finally, if you needed clarification on anything in my post you'd replied to part of, bring it up here
"oh, also ... when you said ... in the last paragraph, were you still talking about ... or had you moved on to ... | I'm asking because if you were still talking about ... it doesn't make sense, but if you'd moved on to ... well, that seems really unlikely too" And then as much discussion that needs to occur, can occur. And it won't take a dozen back and forths
Like the Denon rate about 100watt for 9 Channel. That 900watt combine power max. But your Denon will just consume 600-700 watt from power supply. How can you output be higher than input. And I have not talking about power doubling at lower ohm speaker.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,580
Likes
38,283
Location
Gold Coast, Queensland, Australia
Essentially $20K amps don't have power supplies as good as the ones the Denon amps have because the Denon amps have 5x the number of channels.

It's this type of comment that makes me spit my coffee out.
 

MaxwellsEq

Major Contributor
Joined
Aug 18, 2020
Messages
1,628
Likes
2,427
How can this problem be fixed?
Is it just that the transistors we have are too small and all they have to do is develop a bigger chunk of semiconductor when manufacturing transistors so that their theoretical maximum current is 150A? And even though there's no way the chip could EVER dissipate 150A, the special transistor could be used solo for the "push" or "pull" of an approx 100-200W into 8 ohm amplifier? Basically treated like a single 15A transistor would be and often is - typically operated in an amplifier to an average of 2A with transients peaking at 4-5A (8 ohm
It's not a problem that needs fixing. The manufacturer has made a decision on compromises and concluded the design is good enough. They are able to make an adequate profit for the class of product.
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
It's this type of comment that makes me spit my coffee out.

OK, so the filtering might not be there... The purpose of stating how powerful the power supply is was to show beyond a doubt that the distortion that starts at 12W is not at all possibly from a power supply issue.

Stop making a mess, coffee belongs in your mouth, not the floor!

Here: a power supply designed for 11x 150W channels that's been tested putting over 120W into five at once? It's going to be more powerful than any $20K 150W stereo amp.
Wanna know why?

Because $20K amplifiers are designed properly, and an extremely oversized power supply creates its own problems. Also, there's literally no point in being able to supply more power than the rails directly into 8 ohms would dissipate...
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
It's not a problem that needs fixing. The manufacturer has made a decision on compromises and concluded the design is good enough. They are able to make an adequate profit for the class of product.

Oh. My. God.

Does nobody read?

I used the Denons as an EXAMPLE OF MODERN PUSH-PULL AMPLIFIERS AND A SHORTCOMING COMMON TO A LOT OF OTHER MODERN PUSH-PULL AMPLIFIERS IN THE $200-$5000 RANGE.

The point of the thread is to identify the cause of the shortcoming and have an easy way to fix DIY projects or possibly refurbishments.

99.9% sure the cause has been identified (I identified it in OP and Post 9). The rest is possible solutions that aren't ever increasingly complex feedback circuits putting more and more into the signal path.

If a big transistor can be made that's electrically 150A and thermally able to dissipate just 5A continuous, there may be a way to fix the problem of most modern affordable class ab amplifiers.

How? I already typed a lot. Read
 
OP
M

mike7877

Addicted to Fun and Learning
Joined
Aug 5, 2021
Messages
631
Likes
127
Like the Denon rate about 100watt for 9 Channel. That 900watt combine power max. But your Denon will just consume 600-700 watt from power supply. How can you output be higher than input. And I have not talking about power doubling at lower ohm speaker.

Oh, you're asking a question.

If you hook up quality large powerful speakers to a 4 pairs of outputs on your 3700, and do multichannel stereo playback and then send a tone of like 140Hz to the receiver and then turn the volume on the receiver up, it'll draw more than 600-700W from the wall.

The reason you probably don't pull more than that because if the peak of music is at about 140 watts (which is what the x3700h is capable of 1-3V before clipping), the average power of rock music is only about 20 watts

I hooked up the oscilloscope to my speakers and turned up the volume of rock music until clipping was reached, then I backed it down so that the peaks were 1-3V below clipping. The average power rock music can reach at this level is 15-20 watts depending on the song (calculated by my oscilloscope every 400ms)

Older rock - that'll drop to 12-15 watts. I didn't play anything with a lot of bass loudly, but I have looked at hip hop on my scope and relative to rock's 15-20, it's 25-30.

People think a lot more power is flowing out of their stuff than there is. Probably by a factor of 10.

Also, people greatly underestimate the peak voltage of their 100W amps etc. IMO



I should add: do not draw more than 350Wcontinuous from the wall with your X3700H. As I said, if you keep music just below clipping, even the worst offender is only 30W average. Put that across 9 channels and you have 270W, make that 70% efficient and add 40 watts for other functions and you're at 400W. Now, songs vary in volume, so if you reduce this by 1-2dB, realistically you're at 300W. So unless you're causing a lot of distortion and making way too much noise that doesn't sound musical, there is no reason more than 350W average over the span of hours would ever happen.

600-700W the transformer will overheat in 30-90 minutes from cold. I don't know if there is thermal protection on the transformer - there might be. But if it goes off it's possible something bad happens like a service error - where the unit stops working until you have it serviced because it thinks it's broken because other than a tone generator into multiple speakers, it's next to impossible to draw that much power outside of fault conditions. It's entirely possible the Denon doesn't have fatal errors like some products do as well.

Edit: if you leave 600-700W running for too long, damage begins suddenly and progresses quickly. You'll notice a smell like the usual smell of the stereo, but much stronger and all at once. This is the insulation on the transformer windings evaporating - vaporizing. At a certain temperature (165-185-205, it varies and the exact value doesn't matter) it is no longer stable. When too much evaporates from two windings that are facing each other, a short develops. At this point the transformer is broken. It could be repaired, but the $100-150 it costs to replace is the better path. Instead of days struggling only to end up with a torn apart and shorted transformer, you'll just have to replace it. I'm sure it's not easily servicable like 99% of everything these days, so it's best to just not exceed 450 average from the wall for 1 hour, 400 from the wall for 2 and 350 indefinitely (preferably 300 or less).
 
Last edited:
Status
Not open for further replies.
Top Bottom