Power amp under development

A Concurrent Version System, a software to manage different versions of a project.

The whole diyaudio.com site could use such an integration, imagine how much search effort and redundant discussion could be saved from having a self-containing project management system!
 
Hi gents,

I am very well versed in what CVS or SVN is. This thread or the whole diyAudio site does have alot of duplicate and valuable information or would be far easier to use and reduce the duplicate questions (aka high S/N ratio) in a thread project specific or the more generic information that is common to many projects.

The reality is CVS, SVN or any other tool will not deal with the need suggested by CVS or SVN. The reason boils down to many humans and many humans organizing in their own "unique" ways. The only hope this CVS like idea would work is if one person with excellent human organizating engineering skills organize dand kept it all organized. It would be obvious from the prior statement that is not possible. I have been in IT since the days of Core Memory, as in real iron core memory and I know how It works or more correctly does not work.

I fully agree with the motivation to suggest CVS or similar to address the clearly obvious problems. Sadly any tool and human organization engineering aside, there is a massive effort to scope and then implement whilst the diyAudio site continues to function. That means the conversion and transistion elements are a massive matter in themselves.

With respect to the comments of this thread, I have excellent notes and was going to eit the entire contents of this thread into a Guide that is specific to this thread's amplifier project. I just have not had the time to do so. My guess with my very familar knowledge of this thread is it woudl take me at least 2-3 months of fairly solid effort. Once done then the incremental updates to the Guide would of course be faily easy even on a once a week basis as it would also reduce the duplicate questions that keep appearing. I have some very challenging life matters that have been ongoing since the summer so the work on the Guide has not moved forward after my rough outline document I created late last year and into early part of this year. I do not forsee me being able to pick up the work effort for a few months yet as much as I really like to dig my teeth into the Guide which would basically be the living book of this thread with indexes and cross references in the document as well as a table of contents.

That said, just think of the work involved in doing the same with many oif the otehr threads, not to mention distilling the information into common subtopis that can be referenced by all projects when say need to know about Feedback, PSU, passive filters, gainclones, different topologies and sub classes in each, et al. The guide I wish to create requires the same work as the CVS approach.

The CVS would not solve many of the issues in as far as content, copyrights and all aside. The change management and cross referencing needs to the information on the diyAudio site are very complex not because of the site software used for the forums, but by virtue of fact you are tying to organize human communications. I do not think you will find any software that can organize human communications, let alone the massive conversion, implementation, and education required in hopes everyone will use the site according to the "guidlines" to ensure all stays well after any such conversion/implementation. I have seen far too many databases data to know and basically what is being sugested is a very structured and organized database in which CVS or similar woudl the the overseeing tool. Do all the safety features of a car ensure everyone drives safe and that there are no accidents? No. The same applies to trying to use any software tool(s) to structure human communications. It is simply not possible.

Notwithstanding, it was with excellent and 100% best of intents that the suggestion was made to put the thread or all of diyAudio content in some sort of versioning software tool. As well it is wonderful to offer to do the grount work to do so. I for one wish there was a solution. If there is a solution then please speak up.

Regards,

John L. Males
Willowdale, Ontario
Canada
22 November 2006 17:29
 
hi

Sure integrating cvs into an existing website and sorting out the databases would require a bit of coding but the bulk of the code base itself can be used freely and has been implemented for php and iis.

On the organizational side it would be down to the project owners to set up and maintain their own repositories. Im not sure what copyright problems might arise.

Unless you want many contributors to be able modify the designs as in the case of os software, a sufficient solution would be to have a files section attatched to each thread that the thread owner can edit, or just stick the schematics and maybe revision notes on an ftp server.

Though this was not really what i was really suggesting at all, maintaining a site and making these kind of changes is no small task. I was simply suggesting setting up a project on an existing web-based cvs (or rather hoping quasi might be inclinded to do so) and upload the plans of his fantastic amp.

Definately a condenced version/guide would be good or if you can mail me the info that you have collected i'd be most gratefull.

I dont know much about audio systems yet and this amp is well beyond my current understanding of electronics but it looks a good starting place for a punchy little soundsystem!
 
O.K. I have just built 2 channels last night with 3 pairs of IRFP 450s. I have not tested the full blown version yet, but both boards were tested with a single pair of OPs on +-42 volt supply. I used IRFP 450 pair on one board and IRFP 460 on the other. For the single pair OP stage, I preferred the latter for its overall tonal balance and darker background. More on the sound quality after listening to the 3 pair version running off +-63 volts with a 750 VA torroid and 37,500uf capacitance per rail.

I used my own PCB layout and included the Leach Clip Circuit but left out the DC detection part, as I have a different protection board comprising of softstart, delayed speaker connection and dc detection, all in one.
 
Sounds good

Samuel Jayaraj said:
I used my own PCB layout and included the Leach Clip Circuit but left out the DC detection part, as I have a different protection board comprising of softstart, delayed speaker connection and dc detection, all in one.

Sam,

Can you post some pics? Also Can you share your protection, softstart and delay combonation schematic/board? :) Sounds like something I may be able to incorporate into another project.

Cheers,

Shawn.
 
Re: short circuit protection

davorinjo said:
I was wondering, could I use 35v 22uF bipolars instead of the 50v shown and could a 12v reley be used instead of the 24v reley, lowering the supply to 12v naturally?
This case concerning the protection implemented on Quasi's board.


pejinm said:
Hi Davorinjo,

The protection circuit on Quasi board is NOT short circuit protection. It's a DC protection. You can use 12V (like I did) but you have to change 470uF capacitor to 220uF, this will effect the response.

Regards,
Miodrag Pejin...


As Miodrag confirms this is only a DC detect and protection circuit and does not offer any short circuit protection.

The maximum voltage across the first filter capacitor is 2/3 the rail voltage and the maximum voltage across the second is 1/3 the rail voltage. So a 35v capacitor can be used if the rail voltages are less than +/- 53 volts. You can use a 12v relay as Miodrag suggests by putting a resistor in series with the relay coil. The value of the resistor should be about the same as the relay coil resistance.

Cheers
 
Hi,
VI limiting, whether single slope or multislope should be perfectly possible.

Consider the two halves of the protection circuit as separate mechanisms.

The Top half of the amp is protected by a conventional half VI limiter shorting the drive signal to output rail.

The lower half of the amp is protected by a conventional half VI limiter shorting the drive signal to negative rail.

Many CFPs do it this way.
 
AndrewT said:
Hi,
I hope that was a mis-type!!
2 * 42Vac will give about +-61Vdc when nominal mains voltage is available and rise by upto 10% over that giving a maximum of about +-68Vdc.
You are insane thinking about using 50V electrolytics.


I think he meant noting as in note - ing. The 50v electros are to do with the DC detect input filter.

Cheers
Q