Best t-amp capacitors?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I want to change the following caps on my t-amp:

C3,C4 input coupling caps
c10 power supply?
caps on loudspeaker outputs.

These have already been replaced with the following capacitors.

cheap 4.3uF input coupling caps
cheap 2200uf power supply? (c10)
5% tolerence 0.1uF polyprop accross outputs

The changes I have made so far have made a fundamental improvement to the sound. Bass weight, dynamics, midrange clarity etc. much improved.

I have a second t-amp in need of tweeking... What would the best possible caps be? Please let me know if there's anything else easy I can do to improve things further. I want to keep it it it's cute case, as I will build a charlize at some point soon.

Cheers for all the help so far.
 
Administrator
Joined 2004
Paid Member
This is an important subject, changing the caps on the T-amp can make worlds of difference.

Use good quality film caps on the inputs. Blackgate also work well and are very small. 2.2uF is as large as you should go. If your speakers don't play down below about 40Hz, you can use 1uF.

So far, there has been no "definitive" input cap chosen.
Generally speaking, you don't want a cap in this position that is known to be "bright".

How did you get 2200uF in place of C10? Did you solder it to the bottom of the board? For C10, I like Panasonic FM series 680uF. They are hard to find in the UK, so try FC series.

There hasn't been much work done with the terminal output caps. More needs to be done here. These caps are to attenuate RF on the speaker lines. ~600K-3MHz

I'd love to see this thread expand. The caps in this amp can really change the sound, it would be nice to hear what others have found. So far the results are scattered around the forum.

Post your thoughts here!
 
Ex-Moderator
Joined 2005
I'm building 4 amp3's which I'll be using to test a couple different types of input caps. I've got 2.7uF Sonicaps, 2.7uF Auricaps on the way, and will probably also audition the supplied 3.3uF Panasonic FC electrolytics as well. I have many hours of listening with the supplied 2.2uF ceramic caps and they sound just fine to me.

The first amp3 kits from Jan used 25V 330uF Nichicon HE supply caps on board. The latest kits come with 25V 470uF Panasonic FM supply caps. Don't know how that will affect the sound, if at all, but the FMs should have slightly less ESR than the HEs. I'm sure Panasonic FCs would also work well here although they do cost a little more.

I haven't tinkered with the caps in the output filter because the on board surface mount caps do a good job of filtering and take up much less space than a quality propylene cap. The amp3 also uses 0.1uF differential output caps right next to the outputs on the PCB so there really isn't a need for one across the binding posts, although you don't have to install the ceramic caps on the PCB and could use a polypro across the outputs instead.

The thing about modding these little guys is that a pair of good quality caps can cost more than the kits themselves! Plus they sound damn good "stock" which it makes it even harder to justify the price...

Edit: It looks like this will be a popular topic :cool:
 
The 2200uF cap is in one of the battery compartments ;-)

I actually connected it in parallel with the existing c10 because I was worried about moving the cap so far from the chip.

Please advise me what would be best.

Here's a weird observation:

I got good 5% polyprop 0.1uF caps to replace those on the speaker outputs. The ones I bought were huge (too big to fit inside the amp) so I connected them in parallel with my speaker leads, at the spring clip outputs.

I then tried the caps in parallel with the speaker leads but at the speaker end. I swear it sounded better! Just a bit sweeter...

I'm not jumping to conclusion though - the speakers have good screw down terminals that definitely make a better connection than at the amp.

Let me know if i'm going mad. I have a naim cdp into tweeked t-amp into loth x bs1's.
 
I have just finished numbering the parts on the T Amp board, think it's right?
 

Attachments

  • t amp cct.jpg
    t amp cct.jpg
    98.2 KB · Views: 5,217
Just an idee :xeye:
On my Charlize, is it possible to simply remove the inputs caps? Don't be affraid:hot:
I think about it because my tube préamp has 2.7µF output caps.
Inputs caps on charlize are here to eliminate DC voltage isnt' it?
If the outputs preamp have caps it's coudn't be a problem?

What another caps could be changing on charlize?
 
edge540 said:
Just an idee :xeye:
On my Charlize, is it possible to simply remove the inputs caps? Don't be affraid:hot:
I think about it because my tube préamp has 2.7µF output caps.
Inputs caps on charlize are here to eliminate DC voltage isnt' it?
If the outputs preamp have caps it's coudn't be a problem?

What another caps could be changing on charlize?


I believe the imput caps are there because the TA2020 chip inherently has a DC bias on the input. Therefore the input caps are NEEDED, not to block DC form going in, but to block it from "going out", so to speak.
 
Administrator
Joined 2004
Paid Member
BWRX said:
I'm building 4 amp3's which I'll be using to test a couple different types of input caps.

Good project, Brian! I have been meaning to do the same thing for months, might get around to it soon. I will use the same amp, but with a multi-pole switch to test different caps. Should be interesting.

I gets lots of email on the subject.
 
Hello,
On my ta2020 demo board I use 100uF oscon's for local psu decoupling and three paralleled 1500uF panasonic FA series a little further back.

I use 4.7uF Black Gate N types on the inputs (because I had them already and they fit nicely).

Sounds great.


regards,
PSz.
 
Well for what it's worth, I've put two 1000uF Nic NRSK's directly decoupling C11 & 12 and took the original off the board, just behind that are 4 x 2200 FM's in parallel before my stock Maplin PSU. Quiet as a mouse!
The input caps are Obliggado 2.2's with 36K's taking them in just before the pins (I also moved one of the feedback resistors closer).
The sound is excellent but I'm running in a set of FE207E's and trying to get the balance right. Dificult to say exactly but the MLTL's are giving me plenty of bass and the soundstage seems to be opening up every time I listen to them.
Now if I can just get round to painting the cabs and boxing up the T I'll be onto a winner!

Lee
 
Thanks guys.

Lostcause: Your experience is greatly appreciated - I think I will be using the next t-amp with fe207e's in a voigt pipe.

Where did you get the obbligado's in the UK? I'm happy to order online, if required.

What would be the problem with increasing c10 above the 2200uF I'm currently using? Lostcause is using 4 in parrallel, if I understand. I am no engineer, so I am attracted by the idea of having lots of latent energy availible.

What about the effect of increasing c3 and c4 (input coupling caps)? I know the stock value is deliberately under-spec'd to attenuate bass (to make the most of the amps 6W output) but how high would be worth going?

Unless I hear cries of "NO!" I am happy to go overboard and report back. This is the advantage of such a cheap amp....
 
Administrator
Joined 2004
Paid Member
sharpi31 said:
What would be the problem with increasing c10 above the 2200uF

I don't see any problem at all, but there are those here who would disagree. My approach is to get a mid-size very low ESR cap on the board, with a large stiffner cap offboard. I like the way it sounds, more dynamic, tighter bass, wider/deeper sound stage. But there are a few guys on the forum who say the opposite is true. It's easy to test to see what you like.

What about the effect of increasing c3 and c4 (input coupling caps)?

You can, but there would be several disadvantages:
* The larger the input cap, the larger the power-on thump. Above 2.2uF it can get really nasty.
* In general, you want to go thru as little cap as possible. We all know that coupling caps effect the sound, why use more cap than you need? 2uF is as big as you need, 1uF would work well above 40Hz.
* Cost and size. Easy to understand.
 
sharpi31 said:
Thanks guys.

Lostcause: Your experience is greatly appreciated - I think I will be using the next t-amp with fe207e's in a voigt pipe.

Where did you get the obbligado's in the UK? I'm happy to order online, if required.
Next T amp? good luck, I can't even finish this one!
and give me time, I'm a bit of a novice at this lark!
I got the obbligato's from here:(note, I spelt it wrong..doh!)
http://www.diyhifisupply.com/
Just paid the price on the box!
What would be the problem with increasing c10 above the 2200uF I'm currently using? Lostcause is using 4 in parrallel, if I understand. I am no engineer, so I am attracted by the idea of having lots of latent energy availible

I have got a BHC AEROVOX 10,000 uf (RS or Farnell) that Michael recomended to try next, I'm just having a play! To be honest I've done far too many mods all at once and I will have no idea which ones to rate the highest but the power and inputs are certainly going to be top of the list. (I've now started on the PS1 thing...all the gear no idea!)...and there is no way I'm going to finish any of it before Chrimbo if the wife gets her way..
:rolleyes: :grumpy: :wave2s:
Happy modding

Lee
 
Administrator
Joined 2004
Paid Member
I ordered a pair of the Obbligato aluminum caps and a pair of the oil/film caps to try in my upcoming "Great Input Cap Shootout."

Guess I'll try those, 2 types of Blackgate, Sonicaps, AudioCap Theta, Pansonic FM, Panasonic film, cheap electrolytic, and there will be 2 slots open for other types.

Any requests? Ideas?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.