Hello Dadod, I would also recommend reviewing the Swedish company Lab.Gruppen Class TD® amplifiers.
Here are some service schematics:
Free Labgruppen Diagrams, Schematics, Service Manuals :: Schematics Unlimited
And their patent USPTO #5,200,711
US Patent Full-Text Database Number Search
From their service schematics fP2200 they mention the use of flyback converts, a bit different topology.
Good luck with an interesting project!
It was already suggested in post #55.
Hello Dadod, I would also recommend reviewing the Swedish company Lab.Gruppen Class TD® amplifiers.
Here are some service schematics:
Free Labgruppen Diagrams, Schematics, Service Manuals :: Schematics Unlimited
And their patent USPTO #5,200,711
US Patent Full-Text Database Number Search
From their service schematics fP2200 they mention the use of flyback converts, a bit different topology.
Good luck with an interesting project!
Thanks for the links
Thanks for the links
Thank you too Harry for the courtesy.
Btw, I have to correct my self regarding the previous post after having examined the schematics a bit closer, the fP2200 (incl. fP2400q, fP2600) is not a Class TD topology, but the fP3400 and fP6400 are, and they use fixed frequency PWM buck converters feeding the OP stage, if I got it right this time.
Regards
Attachments
Conceptually this sounds similar to the Feed-Forward error-correction system that Sansui used in the early 1980s. US patent number 4,367,442. US4367442.pdf
Attachments
This was my idea!
The difference is you know how to do it and I don't😂 Please do it and do it good! I would buy one Building a Class D amp acting like Class A. Is it possible?
Hahahaha of course I know there are hundreds who thought about it before me. Mine was just a joke. Idea is something to excute it is something
I had a similar YouTube idea YEARSSS ago but the idea just stayed there Idea is just a small part of it IMO.
Indeed - I think the main difference is the use of a class D amplifier to provide the power.
It will need a bit more thought - I am thinking of ways to simplify the signal summing through Rsum
Design and Construction of a Feedforward Error Correction Amplifier
Takahashi, Susumu; Tanaka, Susumu
AES Convention:65 (February 1980) Paper Number:1614
When we tried to make reverse engineering of AU-D9 we found that the low power amplifier (correction buffer) and ff network were not stuffed The amplifier sounded good though
Takahashi, Susumu; Tanaka, Susumu
AES Convention:65 (February 1980) Paper Number:1614
When we tried to make reverse engineering of AU-D9 we found that the low power amplifier (correction buffer) and ff network were not stuffed
The amplifier sounded good thoughAttachments
Hi David,
Not yet. I am having cataract surgery, one eye finished one to go, so next 30 days I have to be careful with my eyes and then I will start with this project. I hope this winter it will be finished.
Best wishes
Damir
I hope the operation is trouble free and that you are fine.
In the meantime I will think about efficient amplifiers.
Best wishes
David
Hi Harry (if you are still here while Damir recuperates)
I think the same loses would apply to simple buck down-converters, as used by Labgruppen.
They seem to be almost the default choice for professional use so presumably it not too serious.
Have you simulated or calculated this to see how much it difference it makes?
I also noticed the so-called "BASH" amplifiers.
They use a double ended linear amp, so they can just one class-D amp.
Seems clever but I haven't studied it yet.
Have you (or any lurkers) any comments on it?
Best wishes
David
Same as what?
No. I'm afraid the best I can offer is this rather vague observation: it's very tricky to determine what the real-world result will be, because it depends heavily on factors which vary wildly, namely the program material and the load. If this combination gives rise to frequent large phase disparities between output voltage and current, then the impact on losses will be very significant.
It may also cause interesting distortion effects, because when the output voltage and current have the same polarity, the conducting linear-stage output transistor will be operating under constant Vce, but if current and voltage have opposing signs, the output transistor Vce will be varying, increasing distortion.
Do you have a link?
Last edited:
I didn't write very clearly, I meant to compare the losses of a class-D solution, like Damir proposes, to a circuit that only sends the rails down to zero, same as the Yamaha EEE, the Labgruppen or the Carvers I have seen.
I realize it's very dependent on details of the input and the load.
I just wondered if you had a feel for how much difference it made it practice, with realistic speaker loads to shift the phase.
I see the benefits of Damir's approach in terms of constant Vce but perhaps a simpler approach will work as well in practice.
The product was aimed at the OEM market and information is a bit scarce, commercial-in-confidence and all, but I tracked down what seems to be the key patent. US5075634A - Composite bridge amplifier
- Google Patents
Best wishes
David
Excellent, thanks
Hadn't come across that one before; it's an interesting approach.Sorry I couldn't be more helpful with the other question
Here's another question for you, and Markus or others "skilled in the art".
Have you ever considered a Zero Current Switched/ZVS circuit for a class-D amp?
There's been a lot of work on this for SMPS for the usual reason of increased efficiency, but it has the nice bonus of reduced EMI, and potentially faster I suspect.
Seems like it would be a clever idea for class-D but I haven't seen it.
Have I just missed it because class-D isn't my area or is this a step forward?
Best wishes
David
The challenge there is to maintain the zcs or zvs condition for all input/output scenarios, without adversely affecting distortion. Having an auxiliary leg that injects the right amount of current into the switch node is one possible approach but is expensive to implement.