Soongsc: I have so far only tried with traditional power supply, but i guess SMPS will work fine too.
Today i have been working on implementing a new function in the modules. It's a Black Box function. Every minute vital data is stored in the chip, so in case of diagnostics are required, this data for the last 2 hours of operation can be downloaded into Excel, and used for service and support purposes.
Today i have been working on implementing a new function in the modules. It's a Black Box function. Every minute vital data is stored in the chip, so in case of diagnostics are required, this data for the last 2 hours of operation can be downloaded into Excel, and used for service and support purposes.
Lars Clausen said:
Just wondering whether the switching frequency in an SMPS would be of any significance to matching with this module.
Re: Re: Patents
It's so funny!
Anyone, even without any basic of high school, knows that the last stage of the audio power amp is connected direct to the speaker terminals, otherwise or it's not an audio power amp, or it's not a last stage!
In the attached sch. single voltage loop took off NOT at the speaker terminals, but before the Pi-filter. Does this topoligy violate the Philips patent?
Bruno Putzeys said:
It's so funny!
Anyone, even without any basic of high school, knows that the last stage of the audio power amp is connected direct to the speaker terminals, otherwise or it's not an audio power amp, or it's not a last stage!
In the attached sch. single voltage loop took off NOT at the speaker terminals, but before the Pi-filter. Does this topoligy violate the Philips patent?
Attachments
I think it's very clear that taking feedback off the speaker terminals has been around for a long time just for simple concept demonstration. The 0.01ohm in the schematic above is the same as direct wire since lots of wire type would have the same resistance due to real conditions.
There are lots of patents that are out there that won't stand contestant, doing so is just giving lawers money for no really good reason. Since all are trying to reasonably improve designs independently with good will, it's just exciting to see products improve.
Really not worth a whole lot of discussion. It would be more interesting to hear what Lars thinks after his own listening tests.
There are lots of patents that are out there that won't stand contestant, doing so is just giving lawers money for no really good reason. Since all are trying to reasonably improve designs independently with good will, it's just exciting to see products improve.
Really not worth a whole lot of discussion. It would be more interesting to hear what Lars thinks after his own listening tests.
Re: Re: Patents
Not quite true, Bruno. US patent 4,041,411 from 1977 is a self-oscillating amp using only output voltage feedback.
The feedback is actually a UCD type compensation, allthough some smalll hysteresis is used in the forward path. However the hysteresis, could be argued, just has been put in in to speed up things.
Bruno Putzeys said:
Not quite true, Bruno. US patent 4,041,411 from 1977 is a self-oscillating amp using only output voltage feedback.
The feedback is actually a UCD type compensation, allthough some smalll hysteresis is used in the forward path. However the hysteresis, could be argued, just has been put in in to speed up things.
You don't feel a very important difference - in this amp (like in all others) the feedback is taken from the last output stage, and in UcD it's taken from the speaker terminals!!! 😀 😀 😀
Lars Clausen said:
Hi Lars,
Despite your honest disclosure and upfront replies people still seem to have the idea that these modules are "intelligent".
I applaud your efforts in looking into such features, the big question at the end of the day will be... would they be worth it? I guess it depends what you do with it, even if it is just from a marketing perspective.
Speaking of which, storing the monitored variables once every minute seems extremely slow given the nature of the system.
So exactly what is being monitored, current, voltage, temperature? I'm trying to see how could that possibly be made sense of in terms of service or troubleshooting, especially when it's only being captured once a minute in a high speed switching power amp?
Also, will you be sharing this stored data with Google, Yahoo, Microsoft, or the RIAA?
I'm curious as to the clock rate of the Risc MCU, type of memory that's being used to store these black box variables, how often/fast are they being monitored in general use?
Maybe it could be made even more user friendly and certainly DIY friendly by using that extra room in the chip to offer such features as a built in PSU slow start or ...
Edward_L said:
It would probably take you six months and hundreds of thousands of dollars to argue that very point against Philips in court, one of many I'm sure. Certainly the UCD is extremely patent worthy, regardless of which claim Bruno tried to argue himself, I think Philips lawyers may do a better job in that regard, he's an engineer after all.
Soren "US patent 4,041,411" cool patent.
classd4sure said:
I am out travelling this week, but i have time for a short answer. I will come back to all of you once i get home.
The type of memory used to stor this data is: EEPROM with >100.000 rewrites lifetime. The Clock Rate is 9.6 MHz. I only intent to share the data with my self and the customer.
If you have suggestions for using the leftover computer capacity, and connectivity, let me hear. I am open to suggestions if they are possible to implement.
All the best from
Sunny Wales UK 🙂
Lars Clausen said:
I few ideas for additional functionality:
Power on self tests before turning the whole module on might be a nice function depending on how the I/O works.
Manual A/B switching between the two feedback loops so that you can easily compare which kind of feedback you like.
High frequency power supply monitoring to see if the power is stable at vital points.
Implement advisory, caution, warning lamps depending on criticality of problems detected.
PC software to dump data and provide some diagnosis alerts.
soongsc said:
Atta boy... run with it! lol
Thanks for the reply Lars, much appreciated. Seems like you read my mind, it was in fact lifetime I was concerned with, eeprom baaad. It's a game of statistics but under continuous use that doesn't seem like all that long of a life.
It would be interesting if you could capture the data much more frequently under certain near fault conditions only, say enough to recreate at least a cycle or two, if possible, and otherwise it doesn't get used at all. This would improve the chances of a very long and useful life and possibly be far more rewarding in terms of troubleshooting. It could simply be tripped when hitting a certain current threshold of your choice, which also automutes, both to ensure it doesn't get too abused, saves hearing and speaker damage, and for you, ensures a long time of that eeprom. You could also keep a permanent log of each time any of the protections are tripped, including which ones, the higher the count the more it's been abused... but then to keep people happy you're probably just replacing their modules anyway so that'd be useless 🙂
As per other options, it all depends on how much of the chip you're using already. Do ya have any free ports left to play with?
I don't imagine such things as auto switching between feedback modes would be entirely useful beyond the first few minutes, or user diagnosis software to be very useful for long term use.
From the simple to the more complex, clipping/warning indicators, mute, auto mute/on, soft start control output for PSU which auto resets upon loss of mains while muted, (seems like alot of those could be combined features...++++) voice activation/control, thumbprint/retina scan anti theft systems, and maglift i/o control outputs for anti vibration hover system....
Hey, in terms of marketting it seems flashy lights will get ya further than a black box.
I was quite happy with an Apple II computer until the Intel 286 came out and the memory went beyond 640K.😀
I'm sure lots of people can live with less functionality. But most people never learn more than they can imagine possible.
I'm sure lots of people can live with less functionality. But most people never learn more than they can imagine possible.
Charles, no sorry via Cardiff this time. Next time i am i Lndon i will drop you a mail. 🙂
All the best from
Lars
All the best from
Lars
- Status
- Not open for further replies.