TK2050 Extremely successful modifications -- list, pics, review

I'm using a charlize2 (have a TK2050 on order)which has no input caps, which is fine if the downstream kit has them on output. Has anyone tried no input caps, on the TK2050?
I have not tried this, but I plan to... Remember the source and amp have to use decoupled power supplies, otherwise you'll have +2.5v of DC going to the amps inputs - not a good thing!

I'm hoping to try the meanwell with no earth connection, if it works ok, then it should be decoupled from my DCX, and I should be able to do away with the input cap.

You can also use input transformers, which is a nice way of going capless.
Mike - IMO, they are better than a $72 TKD conductive plastic, and certainly better than an ALPS "blue velvet" or RK097. Those are the only three I have tested it against.

It also sounds very similar to using a pair of CMF55 Vishay/Dale 1% 50ppm resistors in the signal path. In other words, I don't think they can be beat for the $.

I really want to test them against the real thing to see what they'd be like. Probably not much difference.
Thanks, once I've found out what impedance I need then I'll order some, I'll be more than happy if they turn out to be better than the Alps blue. :) I guess being stepped they'll have less gang error too.
 
Member
Joined 2005
Paid Member
I'm using a charlize2 (have a TK2050 on order)which has no input caps, which is fine if the downstream kit has them on output. Has anyone tried no input caps, on the TK2050?

You mean the upstream kit (the preamp connecting to your Charlize amp) has output caps, so you would not need the caps in the input of your amp.

Most of my amps have no input cap, as my tube based preamp I use have DC blocking caps on thier output.

I agree, the most expensive caps sound worse than no cap at all. Unless your interested in aquiring a certain sound.
 
Hi john65b, yes upstream (duh!) I meant.

I use a yaquin cd1 tube buffer with Teflon ft-3 output caps 2x 0.47 parallel per channel so just under 1uf per channel to my Charlize2 with no problems and no bass rolloff. As my TK250 is still in transit, so cant try it myself yet, I was wondering why nobody has yet tried it without input caps. I see people opting for 2.2uf, which must be for other reasons other then bass rolloff as I dont have that with 1.0uf.

Have you tried it yourself without input caps, as you do with your other amps john65b?
 
gripracer: I can't believe I didn't think of this before, but you may be able to do away with your input cap...

I'm assuming you do not charge your batteries when using the amp? If so, this means your power source is floating (not earthed) meaning you won't need an input cap. I'm not 100% sure on this however, so hopefully someone else can confirm if I'm right about this.

I plan to run one of these boards from li-ion batteries, and another from a SMPS, and I plan to do away with the input cap on the battery powered board...
MikeHunt79, gripracer.

Thanks for the info on the DACT type stepped attenuators, I use a TKD, if this is better, or even only as good for $10, well...

Couldnt resist so pushed the button on a 50k
No problem, I'll be ordering a few myself once I'm 100% sure on what values I'm going to use. :) I'll be interested to hear your opinion on how it compared to the TKD...
 
The most ridiculous part are the big class D output capacitors attached with wires to the PCB. This filter is intended to provide 12dB/oct attenuation up to 10Mhz, to prevent RF from reaching speaker wires (antennas!), but now it's giving up below 1Mhz due to the parasitic inductance that you added.

Replacing three SMD electrolytics, which are paralleled for reducing inductance and preventing resonances, by a big axial electrolytic, which is the highest inductance type, is also an award winning dumb idea. Inductance is typically 1nH per mm of terminal spacing. The big axial cap is around 20 times more inductive than the three small ones in parallel (which where there for a good reason).

The huge input capacitors are also funny because they act as antennas and pick up all the RF radiated by the rest of the circuit due to the modifications (plus any ambient RF willing to disturb the input op-amps).

Are you saying that using axial electrolytic capacitors in this, or any class D amp is a bad idea? What about radial lead capacitors?

Does this also apply to those who have already modded their own sure tk2050 amp and have replaced the whole output stage on a separate bread board?

Maybe someone else will want to chime in with their own thoughts/opinions on this.
 
MikeHunt79, gripracer.

Thanks for the info on the DACT type stepped attenuators, I use a TKD, if this is better, or even only as good for $10, well...

Couldnt resist so pushed the button on a 50k

you won't be disappointed, I promise.

gripracer: I can't believe I didn't think of this before, but you may be able to do away with your input cap...

I'm assuming you do not charge your batteries when using the amp? If so, this means your power source is floating (not earthed) meaning you won't need an input cap. I'm not 100% sure on this however, so hopefully someone else can confirm if I'm right about this.

I plan to run one of these boards from li-ion batteries, and another from a SMPS, and I plan to do away with the input cap on the battery powered board...

Innnnnteresting....

But people with SMPS supplies don't send an earth from their power supply either? Both SMPS and Battery just use the + and - connection.
 
The Charlize2 is very fussy and wont share the ground with anything else. I use a battery during the day for critical listening. The charger circuit is completly isolated with a switch, and I charge over night. If I want to burn in speakers or caps etc, I switch to my SMPS for the wee hours. It is floating as well. So no issues there. So should be the same wit the sure board, if you use batteries or SMPS..... no input caps!
 
Innnnnteresting....

But people with SMPS supplies don't send an earth from their power supply either? Both SMPS and Battery just use the + and - connection.
Well this depends on what SMPS you are using. On my meanwell supply, the - DC connection is connected to the mains earth pin, meaning the supply is not floating. I have yet to see if the meanwell works without an earth connection however, and I need to check if this is safe to do also.

Other SMPS' may give a floating output, often the brick types with only a 2 pin connector to the mains will give a floating output.

This needs to be checked BEFORE you bypass the input caps, otherwise you'll be sending a lot of DC to the speakers, and I doubt it will do the amp much good either, it's best to err on the side of caution here.
The Charlize2 is very fussy and wont share the ground with anything else. I use a battery during the day for critical listening. The charger circuit is completly isolated with a switch, and I charge over night. If I want to burn in speakers or caps etc, I switch to my SMPS for the wee hours. It is floating as well. So no issues there. So should be the same wit the sure board, if you use batteries or SMPS..... no input caps!
Brilliant, just what I wanted to hear, I think I'm going to try bypassing the input caps on my battery powered board when I get the chance. :)
 
Design has been slightly modified for better performance. 2x 2200uf Fine Gold has been changed to 6x 470uf MUSE KZ.

Complete list for those interested:
4x 10uh 3a coils from arjenhelder on ebay for output coils
4x .22uf 100v MKP2 from mouser for decoupling caps
4x .47uf 400v MKP10 from mouser output caps
6x 470uf 50v Nichicon UKZ from mouser for main power caps
2x 100uf 50v Nichicon UKW from mouser for the two little caps next to the little voltage regulator.

_DSC3620.jpg
 
Design has been slightly modified for better performance.

Hi Grip,
have some questions to you or another members here reading this thread:

1) Why you changed your first konfiguration ? Sound quality or techniqual reason ?

2) Is the 2 x 100 W board still available ? Couldnt find this version presently offered. I see only the 4 x 100 W Version of sureelectronic.