Charlize, my thoughts

I guess I don't get the obvious SMPS improvements in my system for some reason. Last night I played around with a Nemic Lambda 12V switching suppy (for way too long) and the Charlize equipped with air core inductors, larger on-board PS caps and auricaps on the inputs. I couldn't detect any differences between battery power and the switching power supply. I had the same result with the single driver Horns and the PSTL's (coax transmission lines).

I typically have at least 28,200uF (six 4700uF panasonics)capacitance between the battery and the amp, do you think that is the difference in my observation as compared to some here? Also, I left the capacitor bank inline when using the switching suppy as well. I do have a Skynet supply coming as well to try, but the Nemic seems to be good (it's rated at 12V, 3 amps max) and is quiet.

Any thoughts on why I'm not hearing a difference (I usually do tend to perceive small sonic differences, especially with the fostex drivers)?

Regardless of my little dilemma, the Charlize sounds great.
 
Nuuk said:


There is one sure way to find out! I should start a 'Capacitors Anonymous' section on DD for all you addicts. :D

Yes, I got started with a few caps given to me by a guy down the street, and before long, I was hooked.:cannotbe:

No, really I started that cap habit by reading some of the recommendations on the amp3 site where 20,000uF was mentioned as a good place to start.
 
I have had a little time to play with my Charlize. This weekend I did the final chassis and at the same time stuck two blackgate 1000uF@50V Standard caps bypassed with a 0.1 Wima Poly. First feeling is "yuck"!

Maybe its the caps being in the way, maybe its the virgin caps, the capacity too large or small but its still yuck!


I don't think I can take it much longer and will end up putting it back the old way sans caps.
I also bought some 10k uF caps as well but am wondering if its worth the effort.
I also have the smps I bought from Apexjr to try which I have not got around to yet.

I also got a battery charger for the ups batteries but that would mean yet another chassis to build, then again I could leave them exposed and when I feel tired or depressed I could just give them a lick!:D

Andrew
 
The blackgate capacitors take a long time to break in. Give them at least 50 hours to break in. Just leave the amp on for a good two days. Also you don't need 50v 1000uf. 16v 1000uf is all you need. I hope this helps.

So I have heard, I will leave the amp on for the weekend and over to see what if any improvements there are, ime 50 hours is not very long though. As to the value well its what I have in stock, left over from Gainclones.

Truth be told maybe a year of more ago I used some Blackgate non polars as DC blocking caps in an OTL amp , (the expensive ones, 680Uf@35V, paralleled 7 of them per channel) and ended up taking them out and putting back the single elcheapo elctrolytic, got the bass back:D and I played it for many weeks to burn in.

I think they are overrated as are most of the boutique stuff but I had to learn the hard way!
Solid engineering always wins, a p.o.s. circuit with boutique parts is still a p.o.s. circuit.
I only started to try the boutique stuff a few years ago when I discovered some of these forums. Some of it does sound different, better is a matter of opinion. I think much of it is good quality but way overpriced.

Andrew
[
 
Nuuk said:


There is one sure way to find out! I should start a 'Capacitors Anonymous' section on DD for all you addicts. :D

Are you saying in a clever way that I should try the SMPS sans capacitors to realize what the others are hearing? Do you think the caps are deminishing the SMPS's sonic advantages?

I don't mean to ask stupid questions, but I am a bit ignorant regarding power supplies. Thanks for the help...
 
Are you saying in a clever way that I should try the SMPS sans capacitors to realize what the others are hearing?

My first response would be yes.
Thats how I was doing mine and while I would not say it was a night and day difference between the smps and the batteries I think it is a bit more dynamic with the smps. The one I use now is a computer p/s.
I have another to try.

Andrew
 
input cap value

IIRC the tripath datasheet shows the input filter as 20K / 2.2uF which should give a -3dB figure of ~ 3.6Hz.
I am going on the assumption (never a good thing to do but in this case I don't think I am far off) that the Charlize uses a 20K on the input.
The 10Uf on the Charlize would give a - 3dB of ~ 0.8 Hz.

Also, again, the tripath datasheet gives frequency response of 20Hz - 20KHz +/- 0.5dB.

So anything ~2uF should be more than sufficient to play even the lowest stops of your favorite organ:)

Andrew
 
Re: input cap value

Andrewbee said:
IIRC the tripath datasheet shows the input filter as 20K / 2.2uF which should give a -3dB figure of ~ 3.6Hz.
I am going on the assumption (never a good thing to do but in this case I don't think I am far off) that the Charlize uses a 20K on the input.
The 10Uf on the Charlize would give a - 3dB of ~ 0.8 Hz.

Also, again, the tripath datasheet gives frequency response of 20Hz - 20KHz +/- 0.5dB.

So anything ~2uF should be more than sufficient to play even the lowest stops of your favorite organ:)

Andrew

i think you are right !

I've found a schematic of a very similar amp than charlize. I'ts the autocostruire:

http://www.autocostruire.com/audiodigit/class-t-e.pdf

The schematic is at the last page.
We can see a 20k/2.2 µF
;)

I've ordered 3 for friends. It's an interesting product because of hole components. Easy to tweak ;)
 
Administrator
Joined 2004
Paid Member
ohenry said:
It seemed that losing the 10uF input caps and installing something <3.0uF helped with the nice turn-on thump as well...

Well heck yeah! Sorry you missed that one, O'Henry. It should be in the FAQ, need to add it.

The bigger the input cap, the bigger the thump. What happens is that the input cap has to charge up to 2.5V at power on. It charges thru the 22K input resistor - simple RC constant. Until charged, the bias voltage is lower than 2.5V resulting in a large DC offset at the outputs. Changing the size of the cap changes the size of the thump.

The new Super-T amp has a clever mute on power up circuit, so no thump. There are easy soft start circuits discussed in this thread, too.
 
seemed that losing the 10uF input caps and installing something <3.0uF helped with the nice turn-on thump as well..

I would not be surprised, more cap to charge. The fix as shown on www.diyparadise.com is very simple however and works perfectly. Turn to first position and wait a few seconds, maybe 5 or so and then "full on" and presto, no noise!

I must admit that with the stock Elna Cerafine as the input it sounds great. I have not done any component changes to mine nor any modifications to the circuit, it sounds fine as is to me!

Maybe one day in a bout of idleness if I discovered some boutique caps I probably would try them.
I would be very careful of any changes with the board, the traces are small and the chace of lifting a pad is high.

Andrew
 
Administrator
Joined 2004
Paid Member
Re: input cap value

edge540 said:
I've found a schematic of a very similar amp than charlize. I'ts the autocostruire:

A nice kit. But why not get an AMP6 form 41HZ? Same amp, but less cost. Less than 1/2 the price!

Don't know how the Autocostruire sounds, but the AMP6 sounds great and is easy to build. AMP6 has a few circuit improvements that this one does not. Don't know if it helps the sound, tho. They should sound very much the same. Bonus: the AMP6 has a PSU on board!

The Autocostruire documentation sure is a lot better than the 41Hz - nicely done. But in the Autocostruire docs they talk about a “Minimalist assembly” where you leave out the diodes. Yikes!! I can hear the Tripath engineers pull out their hair. Do Not leave out the diodes if you want your chip to live long.

Nice kit, probably sounds great - but you could save 41 Euro with AMP6 and buy better input caps, Noble pot and other goodies. Just add box, connectors and a transformer and you're ready to go!
 
You have right panomaniac. I didn't have a look too close for 41 hz. I am very impressive with charlize and the second one in class T amp review from TNT is the autocostruire. in fact, It was the amp3 who was tested :xeye: . My fingers are too big for surface components :D . But, i see that amp6 has hole components. Too late, 3 autocostruire have been ordered :cannotbe: