As I have made the point in my posts several times, yes the Rotel can be improved by modern caps, improved CCS, and a change in the second stage topology going to TMS rather than DP. I did it to an 850 myself. Seems someone has an issue with that and has to prove he is smarter than everyone else.
It was over 15 years ago and I moved on to designing my own. I recapped it first. Always best to start with what they intended. The docs I made went with it when I sold it. What I remember, a couple caps were electrolytic, so changed to film. Added films to bypass the P/S caps. I changed the constant current sources, used green LEDs if I remember. Textbook LED biased transistor CCS. CASCODE FET/FET modeled better in Spice, but I was going for the 90% fix. ( I may be able to find my old Spice files on an old computer where I compared every CCS I could think of) One of the amps I did I added a regulated power supply for the first and second stage, but that might have been a Hafler. It did not help any I could hear over the improved CCS. I remember in my B&K, I installed star power and ground architecture. Not sure it really helped either as the big change was the input and feedback caps.
I then changed the feedback compensation to Transient Miller. Modeled it in Spice. Basically removing a cap and connecting resistors but the detail I don't remember. My guidance was from the book by Dr. Self. ( hint to do so was from John Curl)
I did put the entire schematic for every one I worked on in LTSpice as a first step. Especially with older amps, you may find a few small things not that clever. Play with the bias currents in the first two stages.
Do realize, it is easy to turn an amp into an oscillator. Getting sufficient stability can be tricky and failing can cost you a lot of outputs. There is a reason my MOSFET amp is Exicon not Hitachi. Use Spice to look for your phase margins.
Also, as I have mentioned, this change modifies the profile for distortion. If not really good tweeters ( About $50 Seas and up) you would do better to stay with the Rotel architecture, just improve it a little. For example, I had some Paradigm Studio 20's for a while. Well respected for the price@ $1200 a pair, but the Rotel was far more listenable. Upgrade to my own speakers, and the TMC amps were clearly more "clear". Now, also respecting Nelson Pass and his disdain for feedback et-al, if you follow Self and apply as much feedback as you can getting great low numbers, you may find it a little less musical. Listening still has it's value. So, think about the "20 dB" improvement before you chase numbers.
Have you read all the old Audio Amateur on POOGE? Many good tips, And some magic (A.K.A. BS)
The Amplifier Design Handbook by Self is a good place to start. Also read all of what Pass, Didden, and Linkwitz have to say. Not a fan of Leach.
Once I was happy with my own amp, I just moved on. Everyone has their preferences. Some still believe in 12W OTL tubes. Some have jumped on the class D bandwagon.
That's about all I can remember. If I can find any old files, I'll jump back in.
I then changed the feedback compensation to Transient Miller. Modeled it in Spice. Basically removing a cap and connecting resistors but the detail I don't remember. My guidance was from the book by Dr. Self. ( hint to do so was from John Curl)
I did put the entire schematic for every one I worked on in LTSpice as a first step. Especially with older amps, you may find a few small things not that clever. Play with the bias currents in the first two stages.
Do realize, it is easy to turn an amp into an oscillator. Getting sufficient stability can be tricky and failing can cost you a lot of outputs. There is a reason my MOSFET amp is Exicon not Hitachi. Use Spice to look for your phase margins.
Also, as I have mentioned, this change modifies the profile for distortion. If not really good tweeters ( About $50 Seas and up) you would do better to stay with the Rotel architecture, just improve it a little. For example, I had some Paradigm Studio 20's for a while. Well respected for the price@ $1200 a pair, but the Rotel was far more listenable. Upgrade to my own speakers, and the TMC amps were clearly more "clear". Now, also respecting Nelson Pass and his disdain for feedback et-al, if you follow Self and apply as much feedback as you can getting great low numbers, you may find it a little less musical. Listening still has it's value. So, think about the "20 dB" improvement before you chase numbers.
Have you read all the old Audio Amateur on POOGE? Many good tips, And some magic (A.K.A. BS)
The Amplifier Design Handbook by Self is a good place to start. Also read all of what Pass, Didden, and Linkwitz have to say. Not a fan of Leach.
Once I was happy with my own amp, I just moved on. Everyone has their preferences. Some still believe in 12W OTL tubes. Some have jumped on the class D bandwagon.
That's about all I can remember. If I can find any old files, I'll jump back in.
I still remember J Curl talking in full contept about D Self, considering himself way better than him.I can hardly believe that he ever gave any advice about reading D Self...By the way...puting a well known schematic in LT Spice won't really solve anything. A big company like Rotel payed tens of very good designers to check with the reality of the circuit too! It looks like you never really worked with a real design team.You think that Rotel's top designers needed any LtSpice to design the best amplifier on the planet if needed? They did what they were payed to do, but one of the top skills of a designer is to be able to make an electronic equipment that can't be modified into a better one unless you change everything in it. Making it into something that suits your ears for a few months is a completely different thing than making it a better design.You need to prove with clear cut measurements and worst case scenatio failure assesements that you really did a better circuit. Improving a circuit's thd by 20 db and making it prone to oscillations is not really useful . There are tons of circuits that simulate perfectly in LtSpice and get on fire in their first 3 seconds after Turn On.
Last edited:
unfortunately...yes ...i did!i would have liked to have more dreams about the designer's world , but all the dreams were crashed by a tough reality check.
Therefor knows everything and so can call others a liar. He should learn how to read a bit better too as he implies a lot of things that were not said. The word "TROLL" comes to mind.
Ten minutes perusing the measurements of name brand components on Audio Science Review should disabuse anyone of this ancient chestnut. They're employees, often contract, like anyone else.
It's the same in every industry and every company, being it a car or washing machine manufacturer has a management board that needs to target every consumer niche with a product tailored to its needs so that the client be persuaded into paying for it.If the clients needs arent adressed they think they are ignored along with their own financial power and leave for another offer.Even if you offer something that's not flexible at all, the client can perceive that as the perfect product for its needs if the price is right.It's simply called cost down technology 🙂
Last edited:
I do not like being called a liar. Sorry.
I have provided my experiences with upgrading an 800 Series Rotel and what I learned about amplifier topology. That is what this thread is about.
I have provided my experiences with upgrading an 800 Series Rotel and what I learned about amplifier topology. That is what this thread is about.
Fair enough.
Personally, I remember being told by my granddad:
"Feeling offended is not a carte blanche for a counterattack - and nor should it be."
This world would probably be in a much better place if more people heeded those wise words.
(And born in 1889, living to be 98, old Claes had definitely seen a bit of human conflict).
Personally, I remember being told by my granddad:
"Feeling offended is not a carte blanche for a counterattack - and nor should it be."
This world would probably be in a much better place if more people heeded those wise words.
(And born in 1889, living to be 98, old Claes had definitely seen a bit of human conflict).
So, I may have read very little of Dunning-Kruger (So ok - uhm, so actually nothing).
So, I owe my expert knowledge on that science from the mini opera performance at the 2017 Ig Nobel Prize Ceremony
Watch the entire three act performance at your peril or just:
The Incompetence Opera - YouTube
So, if you have an intense dislike of marginal psychology and/or opera - you are in safe hands😉 So safe.
So now, I too have moved off topic....So off topic.🙄 So rry.
I didn't say he's a liar, i said i don't believe him, which shouldn't upset him as i'm just a nobody...
Last edited:
Ok, so now please let these issues lie (pardon the pun).
I started this thread as a result of a friendly professional banter and a silly wager over a cup of coffee. But it actually turned out quite ok, and it should stay friendly, informative and constructive.
There are many followers of this blog – each with their personal level of electronic and audio knowledge. But hopefully, everyone (like me) enjoying the continuous learning curve.
This is why I often take the effort to write out the real names of the many acronyms that we readily throw about, and if needed actually explain what they mean.
For someone new to this area there is little educational value to read that going to TMS is better than DP, that doubling drivers is key, etc.
With so much experience and knowledge in here, I think that should come with some level of duty to share our experiences (and inexperiences) more widely. Optimally coupled with a basic drawing of the circuit improvements, some text on the reasons why and maybe test results to back claims up. If not directly, then by a clip from - or a link to the background work.
And to encourage and reply politely to objections and questions – basic as well as advanced.
My God, I start to sound like my old fart of a headmaster
– apologies guys, but it seems like the moderators have abandoned ship so someone has to take the wheel.
Keep modding, discussing and learning!
Best,
Per
I started this thread as a result of a friendly professional banter and a silly wager over a cup of coffee. But it actually turned out quite ok, and it should stay friendly, informative and constructive.
There are many followers of this blog – each with their personal level of electronic and audio knowledge. But hopefully, everyone (like me) enjoying the continuous learning curve.
This is why I often take the effort to write out the real names of the many acronyms that we readily throw about, and if needed actually explain what they mean.
For someone new to this area there is little educational value to read that going to TMS is better than DP, that doubling drivers is key, etc.
With so much experience and knowledge in here, I think that should come with some level of duty to share our experiences (and inexperiences) more widely. Optimally coupled with a basic drawing of the circuit improvements, some text on the reasons why and maybe test results to back claims up. If not directly, then by a clip from - or a link to the background work.
And to encourage and reply politely to objections and questions – basic as well as advanced.
My God, I start to sound like my old fart of a headmaster
– apologies guys, but it seems like the moderators have abandoned ship so someone has to take the wheel.Keep modding, discussing and learning!
Best,
Per