This sounds very good to me, but let me tell you something.
For one thing 1K are definitely not enough to build something decent.
Just the metal case with heatsink will be over $600.
With the left over $400 you will have to fit a very powerful tranformer a hige set of FET to be able to match them and then connectors, filter capacitors and the rest of hardware.
I mean I personally think we are talking about more like 1.5K for one channel for a nice powerful amp.
Beside it is not easy to find very good heatsink that will be able to handle several Ws of heat and that will also fit niceley with my metal enclousure.
So my idea was to get another X250 (I understand it could get a little more expensive...and that is why I will probably start looking around to see if I can do it for less like you suggest) and keep my unit up and running enjoing my music and make one channel and compare it and keep iterating till I get something that fully pleases me.
I mean it is easy from your side having access to all the CNCs but in my case it is a pain just to get a precise set of nice holes on a metal enclousure or heatsink.....and even making one hole it is a pain since I am not equipped for metal work.
is the DSP your own desing? What chip are you using?
When you talked that you learned to program uC do you mean that you learn to write code?
What platform did you use?
I am interested in knowing more details.
What comparisons have you ran? A/B tests, what DAC did you compare it to?
Wooooppyy chill out!!!
I agree with you regarding that retail equipment is marked up, and everybody knows that since the company has to support costs, employes etc...
I am not saying that your thing couldn't be comparable to a 2-3K dollars unit...I am pretty sure it does.
If I don't recall wrong I was talking about reference gears.
This equipment goes way beyond that price!
I wouldn't put a 2K dollars DAC on my system and actually dimonstrate it, not because it is bad, but just because it makes no sense to match it up with a speaker that costs more than 10X and also the rest of it being several time more expensive....anbd it would just end up playing not as nice.
I have an analogue front end that cartridge and table and arm combo is about 12-14K...I can tell you that no matter what you do if you put a cheaper table or cartridge or arm, you will hear it big time....and I have tried it since I had owned cheaper things and upgraded with time.
Not that the cheaper things are unlistenable but they are simply not as good.
If I am a turntable manufacture, do you think that my table would cost me that much? of course not....I know that...everybody knows that....but since I am not a TT man, I go and buy one that fits my quality's requirements!
Same thing goes for my cartridge, arm, cables and so on...
Read my post again. I did not say $1000 inclusive of chassis. I said get 2 chassis then prepare to spend a $1000 in parts.
6 channels of a blameless, CF output stage with 2 pole comp, runs on 42 volt rails. Measures blamelessly too, the implementation took some getting right but I got there in the end.
Currently the multiple subs use the TAS5630, it works surprisingly well. Standard AB amps sound a touch tighter but I've not been able to work on that design for a while now.
Since I've used the FST and the HDS in the wave-guide the amplifier power requirements have been reduced significantly and when I'm able, I'll be making some blameless class A amps for the mids and treble. These will theoretically use a sliding bias, controlled by a micro, to keep the dissipation down when the current isn't needed. This should work well if all goes to plan but often these things have a way of biting you in the **** when you least expect.
It sounds like you have spent over $70,000 in your stereo setup.
If you are comfortable spending that sought of money, I have no idea why you are here
Yes it's my own design using the top line sigma DSP chip from Analog Devices.
Yes and learn how to write the code that configures the micro controllers. This for me was something that I'd put off doing for years as I had no clue about it at all. Then as time progressed, more and more of the interesting chips were all turning out to be software controlled. I decided to spend some money at something I knew I might not end up succeeding at and as learning curves go this one was about as steep as a vertical cliff. The first couple of days that I was working on it were very tough, where nothing really made sense, then something clicked and it began to make sense and I had success in making the micro do what I wanted it to do.
I used a microchip micro controller and programming device, microchip provide you with free software that allows you to write your code in C, which it will then compile for programming the micro controller. Many years ago this used to be a bit of an issue as you'd be writing in a low level language and the programmers would cost a fair amount too. Nowadays though you can get simple USB programmers and starter kits for very little money. Microchip make something called the microstick, which is perfect for beginners to get started, it's cheap, is entirely USB powered and comes with a couple of LEDs on board. That might sound simple, but you can do a lot (from a learning perspective) with an LED as an output. I used the microstick to start with and think it's wonderful for what it is. If you're interested i learning how to program micros then the microstick makes a lot of sense as it's a completely integrated part that works on it's own, ie nothing can go wrong except your bad code.
Well the previous system used PCM1792s in dual mono, these fed into an analogue active crossover that then fed the power amps that fed my loudspeakers. The system basically went from that, to removing the entire analogue active crossover and replacing it with the DSP unit, followed by 4 PCM1792s.
I haven't bought or used any commercial products for years, I just build everything myself, performing measurements as I go to make sure that everything is functioning correctly. If it doesn't it's usually back to the PCB design software for yet another layout.
With regards to the pricing of equipment that you mentioned in the other reply though, when you start making stuff for yourself, you end up realising that the stuff in side most of the incredibly expensive hifi gear, is no different to the stuff in products costing significantly less. Take a well built and well designed $1000 power amplifier and compare it to another amplifier of the same type, (ie same output power, same biasing scheme) but costing $10,000. Does the 10k amp have anything in it that makes it worth the extra? Almost certainly No. The stuff that really matters in power amps is generally the design of the small signal stages and the physical layout. The small signal stages cost like a couple of euros/dollars and as both need a PCB, laying it out correctly should occur in both the 1k and 10k designs. Put simply, there's no reason why the 1k amplifier cannot have the same circuit and PCB as the 10k design. Sure, you could probably improve upon the 1k design by using a larger transformer, larger filter caps and larger heat sinks, but you could do this easily for an extra $500, not an extra 9k.
I am really impressed at the amount of work and effor you put on your project.
This thing ends up being priceless since you developed other skills.
I have good understanding of Microchip's platform and Atmel as well.
I have been designing Embedded circuits and work on software for several years and I know how fun it is to start with a kit and LEDs and see your code shaping out.
I disagree with you about the 1K power amb being the same as a 10K power amp.
There are a lot of details that are not going to be found on a cheap power amp that are on the expensive stuff.
For instance more output stage, bigger transformer more filter caps, more heatsinking.
All these things itself cost very much.
Then stepping up a notch with price, you start finding higher quality PCB, connectors, Transformer and Filter Caps, better semiconductors and quality resistors, and matched semiconductors which as we all know is a big deal.
The quality for these things really, really affect final price.
Then something that is hidden there is the huge amount of work to work out details.
This can be put on cheap design as well, but no manufacturer will put their best circuit on their entry level stuff.
Yeah you can copy and make it for your own, but the things above will certainly cost a ton of money anyway.
Last edited:
I only use Analogue FE as I yet have to find a digital source that can provide same musical qualities.
I am here because I like to build stuff at times since I am an EE with Passion and means for the audio.
In this instance, when I am done with other projects I am working on, I would like to attempt building a power amp and take it to the prong and see what I can make and also make reasonable comparison to grasp the level of implementation I have reached.
I know I have been using their chassis for several years and I think I have mentioned hifi2000 in this thread as well.
Their metal case is not big enough and I don't think that the heatsink will be able to handle 80-100W of pure class A which would likely be my target.
I am now looking at heatsink and other stuff online and I found an interesting thread on this forum that talks about these stuff and there are links to vendors for good heatsink.
But again only thinking of drilling, adjusing, cutting, filing....etc..ect...all the necessary action to make a decent enclousure to fit your heatsink feels really painful.
True that I can find big heatsink for really, really cheap that can handle more heat than a XA200.5 but again making an enclousure out of it is what really worries me.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.