There was a lot of hype about TIM in the 1970's. These days nobody talks about it any more, as it isn't very interesting (being a special case of SID) and because it is very easy to avoid altogether. Are you aware of the work on analyzing negative feedback by Bruno Putzeys and many other contemporary people (as opposed to stuff from 40 years ago)?
Well as his screenshot is of a THD+N test, not any form of TIM test I think the phrase 'all the gear and no idea' springs to mind. I stand by my statement that the 32 tone test is b****y impressive with any grot well below hearing thresholds even if you don't believe in masking.
I wish I could provide boards at that price. Heck, I wish I could buy the raw board area for that price. $2 is a rather small fraction of what I pay my manufacturer to produce the board. $2 is not acceptable for a circuit board of the quality that I'm offering with the engineering and knowhow that went into it and the level of documentation and support that I provide. $2 barely buys a cup of regular black coffee. $2 is basically free. I don't work for free.
I understand that $65 is expensive to some. We've had that discussion earlier (see the first 10-20 pages of this thread). I believe I provide good value for the price. So do my customers. I have received several emails from customers who found my prices quite reasonable. I'm sorry you aren't among them, but if you expect to receive a quality product for free you may have to learn to deal with disappointment - or change your expectations.
I measure the audio parameters on an Audio Precision APx525 analyzer. I do not have the advanced generator option ($2500 add-on as I recall) so I cannot measure TIM. I can measure IMD and multi-tone IMD. I have provided the results both here and on my website.
I measure the loop stability using an HP 3577A network analyzer. For frequencies below what the AP can provide, I use an HP 3563A. For time-domain measurements, I use a TEK 2465B o'scope and HP 3312A function generator.
Tom
I think people have stopped talking about SID and TIM because it's a solved problem. It took a few bad IC and power amp designs before people figured out that if the internal stages were slew-rate limited, bad things would happen to the signal and crappy sound result. Any competent circuit designer should have looked at the performance of all stages in the circuit and not just the combined response.
Bruno does cool stuff. I tip my hat to that guy. He runs a 5th order system with a feedback loop and somehow manages to get it stable. I would like to point out that the Modulus-86 Rev. 2.0 has as much loop gain at 20 kHz as Bruno's amps.
Tom
Indeed. These days it is a non-issue as long as one follows some reasonably clear, simple and well-known design rules. Talking about TIM these days is like talking about accommodating 8-track tapes...
He has made me really appreciate the importance of thoroughly understanding the theory instead of just relying on old wives tales.
While Bruno is definitely a bright guy and good engineer, his contributions are largely into applying solid engineering to specific problems -- such as Class D design. I don't see that he or other 'contemporary' researchers have discovered anything new -- they've just 'done the math' using solid engineering. I.E., I don't see any major discoveries in feedback (control system) theory save perhaps applying to sampled data systems.
I agree - he is more of an engineer than a researcher, and as I wrote, he understands the theory - and has been pushing the envelope of how we understand the theory rather than developing entirely new theories. Most control system theory is pretty well understood and researched - the problem is that it definitely requires understanding the math, something a lot of "audio designers" don't seem to be very good at.
He discovered how to make a self-oscillating Class-D design work subjectively and objectively as well as a classic linear. He then managed to persuade a significant portion of high end manufacturers to use it in their reference $$$$ units. In my book that makes him a bl**dy genius.
In doing that he can now sell things like the Kii-3 which is technology to the nth degree thrown at speakers and the flooby dust brigade at magazines take it seriously. When you look at how many 'well engineered' products have failed in the market place compared to well marketed fluff you release he is a lot more than an Engineer who gets control theory better than most.
IMO YMMV.
I'm sorry I like to joke.
If you design LM3886 THD < 0.02%. I will praise you. Because it is at least 95% true and reliable.😕
But my friend, a Chinese people from the United States told me he is going to use your LM3886
THD + N 0.0004%.
No matter how others think about this question. But from a tested many machines.
I have never seen such a test.
In the actual tests, or even a conductor, or tin, or relay, capacitance of distortion.
More than 0.001%.
I have never done the data looks good, in order to score to design the amplifier.
Anyway, do a low distortion of the amplifier is still worthy of recommendation.
I really have no way to imagine actually under the conditions of high power, such as 50 w?
Its distortion than PCM1794? I'm sorry I still sceptical.
Until one day my own measurements.
In addition, I don't know what you used the servo circuit, DC SERVER in my experiments.
DC SERVER did not reduce the distortion of the amplifier, on the contrary, the DC SERVER is greater than a capacitor distortion,
This can be achieved by many simulation software testing, DC circuit THD SERVER standards will increase, not less.
Hope to have chance to see more tests.
Its distortion than PCM1794? I'm sorry I still sceptical.
Until one day my own measurements.
In addition, I don't know what you used the servo circuit, DC SERVER in my experiments.
DC SERVER did not reduce the distortion of the amplifier, on the contrary, the DC SERVER is greater than a capacitor distortion,
This can be achieved by many simulation software testing, DC circuit THD SERVER standards will increase, not less.
Hope to have chance to see more tests.
No he didn't. It was developed at MIT in the early '90-ties which resulted in the Spectron amplifiers and Tripath Class-T. The concept was taken over by Philips and Bruno refined the concept further. Philips didn't saw market potential and sold it to Hypex. Not a discovery but plain good engineering.
afaik they didn't sell to Hypex, but Hypex was allowed to use the patent (on a feedback loop including the output filter, Bruno's invention) free of charge, on the condition of using NXP parts.
You can argue, indeed. I've heard a Spectron in 1995 and the first Tripath boards a year later. They sounded pretty good 😀 Better than the common HiFi Class-AB.
[Edit]
Yes, that's right vacuphile, thanks for the correction. But I don't consider that as an invention but as a development.
[Edit]
Yes, that's right vacuphile, thanks for the correction. But I don't consider that as an invention but as a development.
Last edited:
A patent was granted, therefore there was a development that was considered 'novel' and protectable above and beyond prior art.
But you are the only one to have used the word 'invention'. Not sure what your problem is. Everyone I know who knows their onions is impressed with what Bruno has achieved. You clearly have a downer on him. All of this is badly off topic to Tom's vendor thread.
Read over the post of vacuphile I was referring to.
B.t.w. Going personal is a poor way of debating. Please keep it on the subject.
If he has a patent then it can be seen as an invention. I missed vacuphile noting it. But unless you can find prior art the patent remains as proof of novelty.
I am not going personal. If I was I would have deployed Troll technique 63 'so what patents do you have'. But you seem to have something against Bruno so I was asking what it was. As for keeping it on subject.. you are the one moving it off subject. Note this is a vendors forum thread for Tom?
I am not going personal. If I was I would have deployed Troll technique 63 'so what patents do you have'. But you seem to have something against Bruno so I was asking what it was. As for keeping it on subject.. you are the one moving it off subject. Note this is a vendors forum thread for Tom?
It is an invention atm because it was clever and nobody had done it before, that is, including the output filters in the feedback loop. This is not trivial because of the phase shift of the output filter in the pass band. It has to be compensated for and I consider it a clever invention to find a way of doing that.