Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Class D Switching Power Amplifiers and Power D/A conversion

Purifi 1ET400A Measurements
Purifi 1ET400A Measurements
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 1st August 2019, 02:54 PM   #11
MorbidFractal is offline MorbidFractal
diyAudio Member
 
Join Date: Nov 2017
Purifi Patents. EP Register might be more useful but the following might not be complete.

About this file - European Patent Register
About this file - European Patent Register
About this file - European Patent Register

The "All Documents" list may be of further interest. My read is that these patents are going to fail due to prior art, lack of inventive step and more importantly consideration of how someone 'skilled in the art' would approach the 'problem to be solved'. That seems to be a general theme in the written opinions.

Perhaps something might be salvaged but I would be slightly upset if someone gained a patent on a generalised compensation scheme and a cookbook of circuit limiters.

Mr Putzeys seems to have already voiced such an opinion in respect of patents and their general utility and of course you have to recognise that he has the analytical recipe to make an edible meal.
Attached Images
File Type: png Screenshot from 2019-08-01 14-25-27.png (39.7 KB, 468 views)
File Type: png Screenshot from 2019-08-01 14-40-34.png (127.9 KB, 471 views)
  Reply With Quote
Old 1st August 2019, 04:13 PM   #12
yoodog is offline yoodog
diyAudio Member
 
Join Date: Feb 2019
Quote:
Originally Posted by MorbidFractal View Post
I would not wish to be misinterpreted if I stepped in and gave a shout for Putzeys but 'wondering where he gets his research from' makes it sound like he is ripping someone else off. I often rip off ohms law but sometimes have to be slightly more insightful when I combine it with other stuff.

As for 'esoteric control theory originating from Russia' and the unknown designer working on marine military radars I'm sure the marine mammal fraternity think he is a bit of a dick and ISTR the original designer at NuForce worked on Tomohawk Missiles and managed to produce something that was unhappy with certain loads.

Having said as much it might be better to properly link to a product rather than leaving us to guess. Perhaps yoodog can consolidate his cherry with his second post to the forum.

Iím in contact with the developer and with person who has installed his first gen amps and got first or second places basically in every single or car audio competition in Russia and nearby on them - to be fair SQ is extremely popular in Russian car audio, there are at 4 different international formats that take place in and around Russia (Ukraine, Kazakhstan and some east Eu countries even) + he participated in the only big EU car audio competition Emma and got great results on his system when it was not even fully set.

About developer background. I just shared what his background is, because his head working is something I have not seen in my life - if you ever saw passion and professionalism working together than you know what I mean. I have been waiting for 6 a while for him to complete his v3 now and it seems like he did it finally :-)

I will be happy to share more info once avaible, but that is all I have for now. Oh new amp name is Indra v3.0 as far as I know
  Reply With Quote
Old 2nd August 2019, 04:26 PM   #13
Reactance is offline Reactance  South Africa
diyAudio Member
 
Reactance's Avatar
 
Join Date: Jun 2005
Location: Cape Town
Quote:
Originally Posted by MorbidFractal View Post
Purifi Patents. EP Register might be more useful but the following might not be complete.

About this file - European Patent Register
About this file - European Patent Register
About this file - European Patent Register

The "All Documents" list may be of further interest. My read is that these patents are going to fail due to prior art, lack of inventive step and more importantly consideration of how someone 'skilled in the art' would approach the 'problem to be solved'. That seems to be a general theme in the written opinions.

Perhaps something might be salvaged but I would be slightly upset if someone gained a patent on a generalised compensation scheme and a cookbook of circuit limiters.

Mr Putzeys seems to have already voiced such an opinion in respect of patents and their general utility and of course you have to recognise that he has the analytical recipe to make an edible meal.
Figure 18: Is that a conceptional sample and hold circuit (used as an example analog to digital converter using the switched capacitor technique)
__________________
Thereís nothing worse than the smell of burned output transistors!!
  Reply With Quote
Old 2nd August 2019, 07:39 PM   #14
MorbidFractal is offline MorbidFractal
diyAudio Member
 
Join Date: Nov 2017
Quote:
Originally Posted by Reactance View Post
Figure 18: Is that a conceptional sample and hold circuit (used as an example analog to digital converter using the switched capacitor technique)
More like an integrate and hold. Figure 16) looks like the limiter Sous has used in their implementation. You would have to read through the applications, claims and reports to see whether any of them are suggested or accepted as being novel.
  Reply With Quote
Old 4th August 2019, 05:47 PM   #15
ErnestoFB is offline ErnestoFB  Brazil
diyAudio Member
 
Join Date: Dec 2017
Extremely useful to know about this. My project has THD+N of 0.02 in low power and 0.15% about clipping and I not have more choices to improve my project.
  Reply With Quote
Old 4th August 2019, 07:09 PM   #16
jan.didden is offline jan.didden  Europe
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Westende Resort, BE coast
Has anybody deciphered those claims? How are the compensator network, gain stage and primary amplifier actually connected?

Register Plus PDF viewer

Then there is an 'operating mode' with the gain stage disconnected??

Jan
__________________
Cable: a potential source of trouble interconnecting two other potential sources of trouble - Erik Margan
Linear Audio articles - SilentSwitcher - Treg high voltage regulator - High-voltage delay unit
  Reply With Quote
Old 4th August 2019, 08:37 PM   #17
ErnestoFB is offline ErnestoFB  Brazil
diyAudio Member
 
Join Date: Dec 2017
It acts by compensating for the nonlinearities that normally occur in a first order delta sigma converter. However, I believe that only with the electrical scheme to know where and how these circuits work
  Reply With Quote
Old 4th August 2019, 09:43 PM   #18
Reactance is offline Reactance  South Africa
diyAudio Member
 
Reactance's Avatar
 
Join Date: Jun 2005
Location: Cape Town
I'm not so sure its worth the effort to seek depth into this (maybe for research yes), improvements of this modulator has reached ceiling level THD improvements, settling with an NCore modulator design is satisfactory.

For true technological advancement in this day and age, effort should be spent on other aspects like amplifier resiliency, Example: using a micro-controller to provide sophisticated features like monitoring of the attached speaker's voice coil, health profiling, protocol design i.e sending command frames to do fancy things like lower gain vs temperature, measure amplifier abuse, monitor deterioration of fets and activate early warning alarms ..

IMO its still a missing must have feature in commercial and diy audio that's aims for reliability.
__________________
Thereís nothing worse than the smell of burned output transistors!!

Last edited by Reactance; 4th August 2019 at 09:46 PM.
  Reply With Quote
Old 5th August 2019, 10:41 AM   #19
MorbidFractal is offline MorbidFractal
diyAudio Member
 
Join Date: Nov 2017
Quote:
Originally Posted by jan.didden View Post
Has anybody deciphered those claims? How are the compensator network, gain stage and primary amplifier actually connected?

Register Plus PDF viewer

Then there is an 'operating mode' with the gain stage disconnected??

Jan
I'll be bold and suggest that other than causing your head to implode in an effort to decipher the bulk of it it really is just a cookbook of RC pole/zero networks wrapped around a bunch of amplifiers incorporating various limiters and bypasses dependent and independent of circuit operating modes.

There may be a rhyme and/or reason as to why any particular combination is chosen but it is ultimately just doodling with standard component building blocks. Look at something like average current mode control where switch current is sampled/controlled.

For slope matching the sensed current needs to be integrated. The sensed current comes from the inductor which integrates the switched voltage. You have two integrators, one from the power stage and one from the compensator, which, if both were active at crossover, would result in an unstable system. You introduce a phase bump, RC zero/pole around crossover to the compensator to gain 45 degrees phase margin.

Take any/many SMPS control ICs. You will see that the error amplifiers are designed to saturate/limit just above/below the ramp valley/peak. It's an effort to aid speed of recovery. If you were rolling your own you would start slapping in diodes, zener diodes, transistors around the local error amplifier to achieve the same function.

All these things are known. Practical implementation may/will cause other headaches. According to words elsewhere and indeed in these patents the 'secret sauce' is in the implementation. The loop is apparently third order at some point in its characteristic due to the addition of an extra integrator. Even a second order loop is going to be unhappy if it loses the plot and does not find it again.

As per the white paper elsewhere the circuit with the additional integrator has, unstable, solutions if the loop gain decides to tank which it will when approaching clip. It would appear that is largely down to the modulator gain falling with output level. The chosen solution is to 'knock the integrator out'.

Figure 16) as used by Sous is a limit. Whilst the integration is gone the amplifier is still active in as much as signal is passing through it. Mention is made of transconductance which is the mode in which this limit is acting. It's not a hard, zener/diode clamp so the stage still has a voltage gain. Roll your own version of 16)

Reactance mention Figure 19) which is an integrate and hold. In this case the integrator is taken entirely out of the circuit. It could be bypassed via another route but consider the case when the amplifier is about to or has entered clip/instability.

The integrator output was as a level commensurate with that condition. If you place it in hold then the amplifier will sit just before or at clip. When it comes out of clip the lower frequency, unstable solution, is not present, no integrator. When the amplifier starts operating again, switching above the wrong solution, the integrator is taken out of hold at the, previous, right output level and things carry on.

I am uncertain as to how EPO defines Novel and Inventive Step and it would appear that all 13 claims qualify as Novel. However all 13 claims do not involve an Inventive Step. Novel may be application of known solutions to supposedly new situations but if you are using known solutions there are no Inventive Steps.
Attached Images
File Type: png Screenshot from 2019-08-05 10-25-07.png (12.4 KB, 323 views)
File Type: png Screenshot from 2019-08-05 09-48-03.png (59.3 KB, 317 views)
  Reply With Quote
Old 5th August 2019, 11:04 AM   #20
MorbidFractal is offline MorbidFractal
diyAudio Member
 
Join Date: Nov 2017
More wordy stuff. It is sort of relevant to inventiveness.

We 'know' the amplifier has at least two solutions. One low frequency/unstable and one high frequency/stable. They are valid over a range of frequencies.

Trigger a monostable off the switching frequency. Whilst it is being held in reset, timeout longer than a minimum switching period, the integrator is allowed to operate. If the switching frequency drops below some minimum the monostable is set and places the integrator in hold with the amplifier continuing to operate at a, stable, minimum frequency at or close to clip.

When the switching frequency increases, the amplifier is coming out of clip, the monostable is taken out and the integrator taken out of hold. First order thinking/solution that is almost but not quite there. Probably needs an FPGA to tweak things.

Just random thoughts but it is in the realms of 'someone skilled in the art' from someone who is probably not.
  Reply With Quote

Reply


Purifi 1ET400A MeasurementsHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Purifi Audio by Bruno Putzeys and friends. ErnieM Multi-Way 101 21st September 2019 06:38 PM
FET Measurements with Just A DMM EUVL Pass Labs 23 30th June 2017 09:55 PM
Usefulness of In-Room LF Measurements vs. No Measurements (ref FH) sayrum Full Range 66 19th February 2015 09:44 PM
DAC measurements kinku Digital Line Level 9 4th May 2014 07:47 PM
amp measurements GAK Tubes / Valves 0 11th February 2004 04:29 PM


New To Site? Need Help?

All times are GMT. The time now is 05:43 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki