Goldmund Mods, Improvements, Stability

Attachments

  • BruceC.jpg
    BruceC.jpg
    75.8 KB · Views: 462
Hi C2Cthomas,

at least there are some on the forum that apreciates the whole engineering process. It is amazing how much do nothing cr*p are added to circuits thinking that it is smart.

Since this subject is of interest to you I suppose then you also subscribe to MTBF and mimum component count to perform the specific function amicably.

Kindest regards

Nico

This would depend upon the target goals of the product or service and most important - the desires of the customer. As a young lad I worked as a Chefs assistant (peel potatoes and stir the sauce) and can assure you that ingredients and when they are introduced into the mix are essential!!! I marvel when I see it applied properly to things audio - and sometimes less is more (Ala Pass Labs) - at other times something a bit more complex is called for. To get it to all come together - and hit QC and cost requirements is a wonderful thing to take part in. But remember that Process Improvement is a holistic endeavor and there are other considerations - such as - are you even offering the right product or service in the 1st place.

I had two rules that never failed - 1st listen to your customers (internal and external) and then listen to your team (even in obscure places like shipping and receiving or storage and distribution - do you know how much money can be saved by effective storage and distribution???)

Eh - don't get me going - it is a real love of mine - I would need my own thread (heck - I'd need a website) to explain everything. So enough of this OT - back to subject!!!

BTW - if all that mattered was MTBF and component count then we would all be listening to HTIB systems driven by a 400w peak power boom box amplifiers. In my search for "good enough" I don't require the refinement of a $250,000.00 USD system - but I do seek equipment that I actually enjoy listening to. Sometimes that might be a simple 8wpc tube amp and a fullrange speaker - sometimes something a bit more "complex".

The parts count between a Toyota and a Ferrari are not that many - the results are very different. But then the wants and needs of the customers are different.
 
Last edited:
Why don't layout jumper positions, so some could only populate the capacitor multiplier part, and others the CCSed parallel reg too. So both persuasions can have it the way they like in one board, not to make a reg second board in any case and avoid extra cost. It will also be a nice opportunity to listen with or without the whole thing and compare for those curious, and Kean's reg work not to pass non evaluated.:wrench:

Makes a lot of sense
 
Hi Salas/Melon Head,

my comment was only to enlighten not to criticise, I have been in the electronic R&D and manufacturing industries a long time and have been a practitioner of concurrent development, value engineering, and TQM. While I have learnt a lot, it is still not enough and one encounters new challenges almost every day.

What I was trying to impress is to review your work and think laterally if there is another way, that may be better, cost effective and more reliable. If there is not then you have achieved your basic goal.

This is not about the design, it is about the complete process and it is not being hard on Anthony, I believe he has exceptional talent, but there is more than just talent and this is something that comes with experience and as C2CTHOMAS pointed out is listening not just hearing.

It is exceptionally difficult as an individual to conform with design by committee and please everyone on this committee. But strangely enough some of the technically most advanced organisations develop by committee, a more appropriate term is concurrent engineering and the aim is right first time.

It may appear as if I am having digs at Anthony, but this is not the case at all, I am trying to show that there are many aspects in the holistic process of development.

In my opinion he is the project manager of this project, but could be more susceptible to suggestion and properly test these suggestions, not merely dismiss something because it is not completely understood. A good question is do I follow, or do I understand.

Kindest regards and keep it up Anthony, I have been following your whole thread with genuine interest.

Nico
 
Last edited:
Hi Krisfr,

is this a double layer board, if not you have some dangling connections. If it is double layer may I suggest that you do not use vias, but change layer at the component connection, that way you have a leg connecting both layers as well as solder suck through.
 
Hi Salas/Melon Head,

What I was trying to impress is to review your work and think laterally if there is another way, that may be better, cost effective and more reliable. If there is not then you have achieved your basic goal.

Nico

I am all for keeping things simple. One of my favourite amps is the F5 with just two gain stages and not a capacitor to be seen anywhere in the circuit.

However it has been my experience that using shunt regulators on the front end of an amp or anywhere else brings significant improvements in audio enjoyment. So in this situation I feel the extra complexity is worth it.
 
I can already vouch that the shunt regulator on the front end will bring about a measurable improvement on the original design that only uses a capacitor multiplier and 50 Hz line frequency.

Whether this improvement would actually be audible is the question. I would like to think that it is because I also use a shunt regulator on a new development, but it may be my perception.

You need to listen to it with and without, which is not always practical as so many other things change as well, it is not as simple as flicking a switch.
 
Sorry Krisfr, on the other hand it looks nice and symmetrical. try avoiding right angles in tracks, rounded tracks are good but not always practical, normally 45 deg works well also.

I can attest to this aspect of design - also - where circuit paths meet a 90 degrees I would put in a radius flare to get rid of the "sharp" corner. Also done by putting in some 45 degree angle "fills" in the corner of the 90 degree corner.😎
 
Maybe a prototype should be build with a shunt regulator and another with a capacitor multiplier and do A/B testing then report the findings to the forum. We may be under the impression that we are improving the perceived sound.

I think if one wants the "best sounding amplifier in the world", it rules out a single design approach.
 
🙂I really appreciate your input, I am just trying to get the nodes connected electrically at this juncture, I can and will go back and take out the kinks and or corners. Until this year I have NEVER laid out a board with a computer. I am old enough to have used Red and Blue tape with Black connections. So push and shove me in the right directions. I am thick skinned and not easily intimidated. This is JUST a get the job done so testing can be done project. HOPEFULLY a group buy is in the future, and Alex MM or other expert can shame me a little. It will not take much to do so. I just think of it as fun, and I am enthralled with the Global inputs, sharing and group think that would have been no more than a fantastical dream a few years ago.😀
 
Nico, I am glad for your participation.

Krisfr, nice work! The 107k resistor should be a trimmer.

I thought you guys might be interested, I went to a repair shop about a month ago and they had a Harman/Kardon AVR35 and a Denon AVR588 that they couldn't fix, said I could take them home. The H/K seems to have a processor problem so it will probably end up being "inventory". However I've been able to look at the Denon today and it has ALL outputs showing 47V DC offset, and I've traced the problem down to a failed KIA7815 (I read somewhere this was a common problem, but why would the 7815 fail? cheap chinese part maybe?). Supposedly I can replace the KIA7815 and it'll work again, and I haven't found any other problems.

Can't get the thermal amp simulation to work yet. Not sure if I ever will.

- keantoken
 
Hi Anthony, check if they used a protection diode over the regulator, sometimes designers forget that a capacitor can discharge reversely through the regulator and destroy it.

I don't think the Chinese would be stupid enough to counterfeit an LM78xx, they are a dime a dozen from a reputable manufacturer.
 
I think it would good to find a cheaper source of the power transformer too. The one I am using is 13.29 each. That is a lot for a few milliamps. But lets test and listen before going forward. Does anyone know of a Inexpensive source of transformers?