I'd say, no. Most (all?) mass-market audio products are keenly engineered to a specific mass-market price point. Such products cannot afford to be 'over engineered'. High-end audio gear is another matter. It is targeted at customers of some financial means for whom home music playback is an significant part of their lives.
i loved cassette even if find the right track was a little painful. Growing old and lazy i am devoted to digital and try to make it sound pleasant. But i loved the sound out of cassettes. I still have an old Yamaha cassette player with a good sound.
Again i find convenience in use a very good thing. So i am using a pc these days.
that is my exact point from the beginning. If great sound can be got from simpler designs why not spend more efforts optimizing them ?
Are more complex design less dependent on parts quality ? maybe that would be a good thing for mass production. But we are talking here about unique products
And there is another issue 🙂 that i would like to discuss. I am rambling now ...
Let's take the Nakamichi CA5 minimalist line stage that is in the line of what i like to see ... changing active parts for more powerful ones may i get a small power minimalist integrated with Vgain of 10 like the line stage ?
again i really need very few watts ... a Vgain of 10-15 times max would be more than enough. I am moving towards high efficiency speakers.
And the same of course with the Rotel line stage replacing the output pair with a let's say 150W bjts pair ... can that be transformed in a integrated power amp ? in other sectors it is called scale-up. Scaling up a line preamp to an integrated power amp by changing the active parts used. Of course modifying also the passive parts accordingly.
Another question ... better LOW open loop gain and less FB or HIGH open loop gain and more FB ? or this does not make the least difference ?
this is a very important point i guess 🙄
By the way, i am replying to you here also from the other thread on Rotel's amp modding because the last thing that i want is to derail it
these are very important points that i did not know and thank you very much for that.
Now the decision of more gain stages is more clear to me. Stability is of paramount importance. I can tell you a story. Years ago i wanted to arrange an hybrid integrated amp. I had and maybe i still have a Croft Micro tube preamp (i.e. just a pot followed by a unity gain buffer) and i would like to use only the power stages of a Rotel ra-930ax bypassing its volume control and 1st op-amp. I made the big mistake of replacing the FB resistor to decrease the power amp stage gain ... the amp muted ... i put a hand on the fins and i got burnt in a hard way. Yes i am that stupid. I learnt then that the amp was become unstable. I remember the mains transformer start to vibrate ... what a disaster ... i think i still have the amp's corpse somewhere. Still that idea is very strong on me ... the hybrid solution. Tubed front end and solid state power.
i would like to fix a concept. One thing is production ... another one is reproduction.
I am sick for virtual soundstage Many nice guys explained me that often is the result of some sort of processing ... that does not matter to me. I want a monitoring system in the end.
I want to hear all there is in the recordings. I do not want to add processing on the original processing.
Another thing is EQ to compensate listening room acoustic issues ... that is fine They told me that it could work as a decent band-aid. Much better to procede with an acoustic treatment of the room ... but it is just not feasible ... today at least. Hopefully in the future.
It is a little like some discussions on speakers ... someone see them as a musical instrument. For me this approach is complete wrong. They must reproduce sounds not produce sounds.
For instance the last thing i would like to get is a speaker enclosure acting as a sounding board
I did some test ... the deadliest the cabinet the better the sound ... ever. With the best material to dampen the panels being the very toxic lead. It is a wonderful material to neutralize vibrations. Exceptional. But also very toxic. My love for music is so big that i risked the life managing some lead sheets and glueing them to speakers panels. The result is just astonishing. Even cheap speakers image like champions.
ok you are right .... but to come to an agreement ... let's say that i need a very good 30-40W/4omh amp solution. Solid state of course. What would be your 1st choice ? just to fix some reference. A high quality small amp. Commercial or diy project is the same. Just to understand what is your design preference ...
i am not very interested in high power solutions. The Krell i was thinking more is the KSA50 ... low power high current ... quite insensitive of low ohm loads. And said to sound great.
i would like to recall you that i was talking about a basic line stage ... not a giant amp.
I can agree that the more power you want the more complex the design. A line stage must add very little voltage gain and being able to drive properly a 600 ohm load ... to be on the safe side. I see Stereophile doing measurements also on 600 ohm ... the amp with the lowest Zin i found was my Albarry m408 ... 5kohm/0,5V at its input
this is difficult ... for instance i love diamond buffers a lot ... symmetry is a characteristic of beauty in the end. Actually an high quality pot followed by a buffer could be my final line stage ... i will have always a huge gain in the power amp ... or at least enough for my needs.
it really looks like i have to start simulating basic schematic
For instance given a circuit to know how much will be the open loop gain is very important.
I wonder if also PSRR can be obtained by simulation. That would be fantastic.
However a lot can be done also in the power supply to reduce the residual ripple.
For this i find very useful to look at the schematic of the totl preamps of the early 80's
They are using discrete parts and not fixed voltage regulators like today
I am always talking about preamps of course ... for power amp a transformer > diodes bridge > huge caps and it is done.
i have the feeling that today mass-market products are above all undersized ... I had two 40W/8ohm amps ... one recent with a 100VA transformer and another one 40 years old with a 300VA transformer Say which one sounds better ?
There was just no comparison especially when asked to drive more difficult loads ...
I think that the industry has taken the fashion route ... they take care of the look mostly. What meets the eye ... but not the ear unfortunately.
As mains transformers and supply caps of high grade are very expensive they cut on them brutally
Medium power integrateds of the 80's had often 300W max power consumption or even more ... now i see more 100W ... max 150W
The difference can be heard on full range speakers very easily
I listened recently to an old Pioneer sa750 ... impressive for power 🙄 and not only power
It shows the same power consumption of a recent 3kUSD integrated (much nicer i have to say)
I think this tells one or two things ...
Last edited:
You want to see over engineered take a look at the sonance sonamp 260, this is the most overcomplicated ‘simple’ 2x60 amp I’ve seen. It sounds good until you get to the upper mids, then it gets kindly crunchy.
That's hardly over engineered, if anything it's fairly standard in its topology and design. It may look complicated to those without formal education in electronics / engineering.
Looks like way too much to get the job done to me but like you said I’m about as informal as it gets. Of all my amps it sounds the worst and has since new, it now resides in the closet, one day I’m going to use it as a learning tool to see if I can make it sound any better as it’s of little value.
you may reconsider your thoughts after seeing smth like accuphase a680...
ACCUPHASE A-680 SCH Service Manual download, schematics, eeprom, repair info for electronics experts
: you need to pay me to answer all your questions 🙂
Going with low supply rails high current poses the problem of GM doubling and nonlinearities that arrise on low Z speakers.If you have high sensitivity 8 ohms speakers , you don't need high currents so you're saved with an op-amp driving 25w transistors .I would've advised you to read Rutger's SSA-30 amp pag,e but it's deleted now...Something a bit simillar you can find here: https://www.electroschematics.com/power-amplifier-50w-2n3055-mj2955/
I think Aguillar and Peavey were the first to make bass guitar head amplifiers with valve input and transistor output, but Aguillar db 7xx series scehmatics vanished as well from the internet...i probably have one of the largest schematic collections on one of my laptops, but i need to find the time and look for it.Anyway Mcintosh and many others did hybrid designs...you can look on google for hybrid amplifier schematics.
JRC4558
You might want to try replacing the IC501 JRC4558 op-amp with something a bit faster/better.
The power amp is a bit more than typical but maybe to no avail. I see a lot of amps that reflect ideas that are a bit naïve. Commercial products are made by people doing a job and not by any one who has spent a lot of years studying the "art".
You might want to try replacing the IC501 JRC4558 op-amp with something a bit faster/better.
The power amp is a bit more than typical but maybe to no avail. I see a lot of amps that reflect ideas that are a bit naïve. Commercial products are made by people doing a job and not by any one who has spent a lot of years studying the "art".
Thanks, steveu
I’ll document that so as not to forget......any ideas on what to replace it with?
It’s a well built amp probably worth figuring out as they can be had under $100 all day/everyday.
I’ll document that so as not to forget......any ideas on what to replace it with?
It’s a well built amp probably worth figuring out as they can be had under $100 all day/everyday.
JRC4558
Well, first have a read through
JRC4558, worst op amp EVER....
some of which you will need to take with a grain of salt.
Numbers that come off the top of my head are TL072 and NE5532, but those are also older chips and be sure you get a compatible package. If your amp uses the DIP-8 version then you could replace it with a socket so that you can easily try out different chips.
There is another JRC4558 IC001 that is used for a signal detect that doesn't matter much so be sure you replace the right chip. It looks like this actually puts the amp into standby when there is no audio input but that's unusual so I may be mistaken.
Well, first have a read through
JRC4558, worst op amp EVER....
some of which you will need to take with a grain of salt.
Numbers that come off the top of my head are TL072 and NE5532, but those are also older chips and be sure you get a compatible package. If your amp uses the DIP-8 version then you could replace it with a socket so that you can easily try out different chips.
There is another JRC4558 IC001 that is used for a signal detect that doesn't matter much so be sure you replace the right chip. It looks like this actually puts the amp into standby when there is no audio input but that's unusual so I may be mistaken.
I found the equally old mitsubishi m5218 performing better than 4558 at line level and its 3v/us slew rate isn't bad at all with its output current capability.I called that the mitsubishi secret and that chip started making me suspicious on lots of modern ultra high slew rate op-amps that can't handle 25pf without 40% overshoot...
“Scratchy, harsh, NOISEY, overly midrangey and they just severely lack detail. When comparing them to the NE5532, OPA2228 and OPA2604, the 4558 IMO is absolutely put to shame. Those chips have closer specs to a 12AX7 tube and also sound/perform sooooo much better. When I play arpeggios through a circuit containing a 4558, there is such a lack of detail and raspiness that it would be embarrassing in a live performance.“
This quote from post #1 of ‘worst opamp ever’ explains what that amp sounds like precisely! Sounds like it’s a easy enough swap, I’ll read that thread then narrow one down.
Edit......sorry about the slight hijack ginetto! 😉
This quote from post #1 of ‘worst opamp ever’ explains what that amp sounds like precisely! Sounds like it’s a easy enough swap, I’ll read that thread then narrow one down.
Edit......sorry about the slight hijack ginetto! 😉
Last edited:
Without going over 10 pages here of comments, reviews, and opinions, I will state my own beliefs about "over-engineering" of audio equipment.
Firstly, audio, sounds, are an analog thing, for those who don't know.
We humans hear sounds through our ears in analog....... not digital, nor any other form.
I believe in "purity of sound" if one prefers to hear things as close to actual original form as possible.
This then also includes attempting, minimizing, or eliminating anything that would disturb that original sound from reaching our ears.
In some cases, this becomes more difficult, and requires some complexity to subdue.
What some people do not know, is that with, for instance, transistor equipment, a single transistor has its own little "voice" in the matter.
The same thing goes for vacuum tubes....
So each transistor that the original input/source signal goes through, it becomes "tainted" slightly - regardless of how good that transistor is.
This disruption in many cases is not a "hearable" influence to humans, but can be measured under the "microscope" of capable testing equipment - which for that matter, has its own set of "disurbences".
The less that the signal has to traverse though silicon/germanium junctions, to get to your ears, in my opinion, the better.
Ideally, an amplifier of "a straight wire with amplification" would be terrific, but we haven't gotten there yet....
So, in my own designs, I tend to use fewer junctions leading to the speaker.
Simple, basic, yet quite satisfying results.
Compare an opamp RIAA phono preamp to, say, a simpler 2 or 3 transistor version.
Of course there's far less "junctions" for that tiny signal to wiggle through, and that's my idea of "purity" which works quite well, and had been done for ages in various ways.
Same thing goes for power amps.
Less resistors, caps, transistors to effect the signal.
Those OCL amps from the 1970's to date are one attempt to clean things up.
However, more recently, it seems they've gone full nuts/crazy with designs.
Opamp hell seems to be the trend today.
Yeah, load them opamps up..... that signal now traverses though a hundred junctions to your ears.
How come I can build a nice, simple, capable amp with only maybe 10 or 11 bipolar transistors per channel - from RIAA input to speaker, per channel, and its result is a thoroughly enjoyable listening experience?
And mind you, not just using my ears, but others who've heard the amp in question.
And connected to speakers that are favored and adored by audiophiles.
Firstly, audio, sounds, are an analog thing, for those who don't know.
We humans hear sounds through our ears in analog....... not digital, nor any other form.
I believe in "purity of sound" if one prefers to hear things as close to actual original form as possible.
This then also includes attempting, minimizing, or eliminating anything that would disturb that original sound from reaching our ears.
In some cases, this becomes more difficult, and requires some complexity to subdue.
What some people do not know, is that with, for instance, transistor equipment, a single transistor has its own little "voice" in the matter.
The same thing goes for vacuum tubes....
So each transistor that the original input/source signal goes through, it becomes "tainted" slightly - regardless of how good that transistor is.
This disruption in many cases is not a "hearable" influence to humans, but can be measured under the "microscope" of capable testing equipment - which for that matter, has its own set of "disurbences".
The less that the signal has to traverse though silicon/germanium junctions, to get to your ears, in my opinion, the better.
Ideally, an amplifier of "a straight wire with amplification" would be terrific, but we haven't gotten there yet....
So, in my own designs, I tend to use fewer junctions leading to the speaker.
Simple, basic, yet quite satisfying results.
Compare an opamp RIAA phono preamp to, say, a simpler 2 or 3 transistor version.
Of course there's far less "junctions" for that tiny signal to wiggle through, and that's my idea of "purity" which works quite well, and had been done for ages in various ways.
Same thing goes for power amps.
Less resistors, caps, transistors to effect the signal.
Those OCL amps from the 1970's to date are one attempt to clean things up.
However, more recently, it seems they've gone full nuts/crazy with designs.
Opamp hell seems to be the trend today.
Yeah, load them opamps up..... that signal now traverses though a hundred junctions to your ears.
How come I can build a nice, simple, capable amp with only maybe 10 or 11 bipolar transistors per channel - from RIAA input to speaker, per channel, and its result is a thoroughly enjoyable listening experience?
And mind you, not just using my ears, but others who've heard the amp in question.
And connected to speakers that are favored and adored by audiophiles.
Audioequipment is usually way underengineered. Amplifiers: No highpass, no lowpass, no lowcut, no adjustable loudnessfunction, no strobes.
Turntables: No balanced output, no autoreverse, often not even autolift, no strobes
cd/dvd/bluray: No numerical buttens!!!, some other stuff
For the rest and the above there are tons of other stuff that I dont remember right now.
Otherwise I wish you happy days and happy diying.
Cheers!
Turntables: No balanced output, no autoreverse, often not even autolift, no strobes
cd/dvd/bluray: No numerical buttens!!!, some other stuff
For the rest and the above there are tons of other stuff that I dont remember right now.
Otherwise I wish you happy days and happy diying.
Cheers!
Actually with a darlington an extra (current amplification) stage is introduced, so this increases the complexity. Apart from biasing and crossover distortion which should be taken into account when using them (even more complexity). The VK8060 is a very classical simplistic, not minimalistic design. But their aim was to make an affordable and reliable amplifier for the enthusiastic audio diy-er - everything you make yourself sounds awfully good, objective as we humans are.
Your Rotel circuit in #72 is what you would like to see as discussed earlier in this thread. Keep in mind that a collector-collector (or drain-drain) output is a high impedance node and susceptable for instability and hf-pickup, unless there is full control of feedback and load conditions. A darlington does not make it better (extra gain stage) nor lowers the output impedance. Such circuits are not suitable as drivers for loudspeakers or domestic audio equipment setups. Hence no commercial or succesfull units on the market (I exclude the current driven designs to keep this discussion clear - also it is about simplification of designs).
If you want to use those Q513-Q515 as outputs (in this design it is the VAS), it also must have the capability to drive large (-r) currents. So these chaps must amplify voltage and current, temperature stable, linear beta over a large range and not so noisy. A low early effect is welcome too. If such transistors (or fets or tubes) would exist, this discussion was non existant.
Another and succesfull approach has been proven by Hiraga and Pass (& others): Hiraga uses a Sziklai as VAS and emitter follower with a minimal front stage (the input buffer could be omitted) rendering in a three-stage six transistor (bjt/fet) amplifier without the need for additional temperature or switch on/off electronics. Pass masters the single stage amplifier approach; chunky SIT's and a suitable current load. I own a 8W Hiraga and it sounds magnificent. I'm sure a Pass will sound superb too.
My (*) remark in #26 points to my low impedance triode-alike two transistor (+ three resistors) compound which does exactly the same as a triode tube or a SIT and makes such a desired two stage design possible.
Just checked the topic for updates while typing: I also agree with #96.
I miss casettes too. Often sounds awesome and so easy. Just grab the tape put it in play. No searching, pairing and so on. Less great if one was too lasy to write/overwrite the content label.
Cheers!
Cheers!
Hi ! i tend to associate complexity with high parts count. The more the parts the more complex the design.
For instance a Sanken multiemitter is simpler to me that a parallel of 3 different bjts. By the way i love Sanken immensely. That is the way to do parallel output stages ... for me Maybe with an even bigger chip to help the heat transmission. I have been criticized because they say that i overestimate the importance of components matching. I think that the Sanken concept is very very sane. Also from a thermal point of view.
When i see and amp with Sanken i salivate. 😀
And they are integrated in the end
i see. The fact that they are not that common speaks for itself. My thinking was that to integrate driver and power bjts on the same chip was convenient instead of having two separate parts. I was wrong ... again 🙁
Everything but a masterpiece of an amp. I see. And actually i cannot see them used anywhere these days
Thank you for the kind explanation ... i do not think to have fully understood. But i understand is not the right way. No more discussions on darlington
i see. And what about taking them out completely and connect directly 501 to 517 to the output stage maybe swapping 517 and 519 ? is it really needed the intermediate 513/515 stage ? can this mod work and lead to similar performance in terms of sound ?
Very interesting. What is your opinion on this approach ? and on the Sziklai as VAS ? i know that in the past was much more popular. End of 70s early 80s. A simple line stage example in the attachment.
Very very interesting. Is the schematic available somewhere ? i would love to see it. As i said above i am switching to high efficiency speakers and small power amps ... this would be very good if the sounds is that good.
Is it difficult to build ?
i have one copy badly built of the Pass Bride of Zen and i like it a lot. I just put a pot at the input and not the output because stage gain is quite high around 6 times maybe ? but overall a very interesting design
Actually after having a discussion about smps i would like to use a 48VDC smps to power it and changing some values ... the expert told me that some smps have very low noise in the audio range and very convenient.
I will buy a Meanwell LRS-35-48 for my line stage ideas ...
thank you very much for your very kind and valuable advice. Where i can find the schematic of the a 8W Hiraga amp ? i think that would be a good way to start the audio zen route ...
Kind regards, gino 🙂
Attachments
Last edited: