TDA 1541 problem (ping Jocko Homo)

Status
Not open for further replies.
Jump to Latest
#21
Hi,

If I use axial caps for the ladder decoupling, I've been thinking about how to mount them. (Perhaps thinking too much!)

The board layout is designed for 5mm spacing so an axial cap would have to be mounted vertically with the top lead bent down the side. (Or horizontally with the legs bent underneath but there isn't really space.)

Now, there are two choices.

1. short lead to the DAC.
2. short lead to the ground.

Now the leg will obviously have a resistance (and some inductance). It will also act as an aerial conducting into the ground plain or the DAC but I guess we're not too worried about frequencies in the GHz.

So my question to you all is this - is it better to have the short lead between the cap and the ground or between the cap and the DAC ?

Cheers,

Jon
 
#22
Jon, The axial mounting configuration is something I pondered about myself. At the end I settled for the short lead to the DAC and the long ones I took to earth. I also went a step further: with a capacitance meter and a piece of aluminum foil I found out which of the legs was connected to the outer side of the capacitor foil (that shows a higher capacitance, say 5 pF) . I took that side to earth. Reason: otherwise I thought there would be stray capacitance. Of course this is over-sophisticated and such tweaking cannot be verified.
I saw a non-axial cap that might work better on a print (from the Sovcor eBay store):
http://cgi.ebay.nl/0-1uF-250V-HI-TO...86895142QQihZ009QQcategoryZ4662QQcmdZViewItem
 
#23
hi,


axial mounting configuration is something I pondered about myself. At the end I settled for the short lead to the DAC and the long ones I took to earth. I also went a step further: with a capacitance meter and a piece of aluminum foil I found out which of the legs was connected to the outer side of the capacitor foil (that shows a higher capacitance, say 5 pF) . I took that side to earth. Reason: otherwise I thought there would be stray capacitance. Of course this is over-sophisticated and such tweaking cannot be verified.
Click to expand...



let me to share little opinion.

I don't know is it taklkm for the coupling caps.

if yes, I would like to say triode_al opinions was right, shorter legs & short lead to the DAC and the long ones I took to earth. but there will have several more reason for the coupling caps.

1. use 63V MKP will better than 250V.
2. in my personal opinion, use AC caps will better than DC caps, I don't know why probably was the leakage of ac caps better than DC caps.
3. If really like to use shorter distance, why not use SMD caps!! this will be the shortest length.

4. If u did not care the outlook, pls considerate the the arrangment of caps was up,down,up,down,up,down the PCB. THis method I test was quite good but the outlook was not good.

very sorry if I am wrong.


thx

thomas
 
#24
triode_al,

The axial mounting configuration is something I pondered about myself. At the end I settled for the short lead to the DAC and the long ones I took to earth
Click to expand...

This is what I was thinking of doing but why did you decide to do it this way round ?

with a capacitance meter and a piece of aluminum foil I found out which of the legs was connected to the outer side of the capacitor foil (that shows a higher capacitance, say 5 pF)
Click to expand...

You wrapped the foil around the outside of the capacitor and then measured the capacitance between each leg and the foil ?

Cheers,

Jon
 
#25
Thomas,

I'm not sure about SMT. I did think about using polypropylene sulphide (PPS) - something like the Panasonic ECHU which is available in 2% tolerance. The ESL of a 100nF radial cap gives minimum ESR around 2.8MHz (64fs). Using an SMT cap would probably push this up closer to 10MHz which wouldn't be as effective (assuming the ESR is similar).

If u did not care the outlook, pls considerate the the arrangment of caps was up,down,up,down,up,down the PCB. THis method I test was quite good but the outlook was not good.
Click to expand...

Were you suggesting alternating the capacitors, short lead to ground, short lead to DAC, short lead to ground ... ?

Cheers,

Jon
 
#26
JonHarrison said:

So my question to you all is this - is it better to have the short lead between the cap and the ground or between the cap and the DAC ?
Click to expand...

Let me chuck a big hairy furrball in here 🙂 I've been away from "digital" a while so excuse me if this is a bit yesterday

Youre talking about the capacitance in mm of wire - how far away is your lead from ground? Ok so I AM uneducated but this is at 11mhz ish not GHZ... I have messed arround with my cd50 and to be honest unless you do something REAL bad it doesnt make THAT much difference... you'll know what I mean when you do 😉

there are MUCH more important worries for you, what op-amps are you using?? that's going to matter.... (Also If I remember correctly different 1541's have a different sound not that I can personnally tell my R1 from my A 🙂)

nevertheless I follow Thomas' advice short leads to the DAC... 😀 and ASH_DAC's advice on metal film (I think personal choice - maybe MSB decoup should be different :devilr: ) but these chaps have really been there and done it already :hats off: <which should be a smiley!!

If you can, and I was working on a board on this at one point, might be a good idea to rip out the signal at the SAA7220 and do it yourself therafter!!

If I can Ask Limono, which frankly I'm going to, what are your aims for this player? Do you want to fix / tweak or generally make the best darn player you can?

you really (imho) have a potential gem in a 1541 based player but like SE tube amps you need to have FAITH! (in deference to Jocko as per your title I should point out Burr brown make a chip called the pcm1704!! - jeez sometimes I crack myself up 😀)

Andy
 
#27
oh one more thing

If you havent already, search for TDA1541 on these forums, there is a MAGNIFICENT thread where the discussion about the internal construction of the TDA1541 is discussed (at length...) esp the reason for the decoup caps and input methodology- s'all clever stuff IMHO. OK I gravitate towards the 24bit club nowadays but the data pool for the '1541 is unparalleled.

Andy
 
#28
JonHarrison said:
You wrapped the foil around the outside of the capacitor and then measured the capacitance between each leg and the foil ?
Jon [/B]
Click to expand...

Just around the capacitor & measuring towards te one leg or the other one. The idea is that outside it's all earth-field, like shielding. Its not about induction at all.
albert
 
#29
Hi
My goal is to have player up to date with some mechanical integrity and like most of you I'm on budget.I'm not aiming at SOTA as it requires skills I lack. So I change el caps , rectifiers to schottky ,clock and install discrete opamps .Thats it .The mechanical construction of Mcintosh is god (maybe even very good compared to contemporary standards) It's a classic but CD is not an amp nor tuner.Thank You for all advices and links .Regards.L
 
#31
Jon
I already ordered Hagclock board and net audio Board .I will reclock with Hagerman unit It comes with super reg on the board so In can use CD's opamp PSU . Borbely Jfet I/V converter has its own Jfet based superreg. There is just too many options and when I'm trying to experiment and change options I'm frequently so sloppy that I'm almost afraid to touch anything. It's OK when it comes to assembly but experimenting and constant changes caused me too many Bangs. But did you try that DEEM thing and compare with regular clock?? Regards, L
 
Status
Not open for further replies.
Top Bottom