Sure TK2050 - enhancing the bass output

[ZFRaudio]

Good afternoon all,

I currently own a sure TK2050 2 * 100w board, but am not yet happy with the performance. It is one of the earlier revisions that aren't on the "expanded board".

In particular, the bass output of the amplifier seems "weak" and "lacking"

Is there a modification I can do to the output section or powersupply (Eg. Larger capacitors?) to help with the problem?

Many thanks in advance, I have searched this board - but while there are 100s of modifications, I am not sure which would benefit the lower octaves

many thanks

Zack

[ZFRaudio]

Just to add, I run the board at 4 ohms per channel, so I am thinking that the powersupply is perhaps struggling?

Thanks

Zack

[zerokelvin99]

+1 on this issue

[swkbkk]

I too have one of those earlier TK2050 2 * 100w board and find it lacking in bass. My "temporary" remedy is to use a software equalization from foobar2000 player. Sliding up a little on 110Hz, 77Hz, 55Hz (the left most 3 bands of the 18 bands equalizer DSP plugin of foobar2000) improves the bass for me, but I am wondering if there is anything I can do to the amp board itself to improve its bass response.

I find it comical that I am searching for more bass as I normally do not listen to bass heavy music such as heavy rock, trance, etc. I feel that double bass is rather weak in pieces like Miles Davis' So What or Bill Evan's The Washington Twist. And vocal pieces like Katrine Madsen's rendition of Close To You doesn't sound full. I hope you get what I am trying to say.

I have done most of the modifications suggested in the other forum. The major ones are change of input caps to 4.7uF Bennic XPP (is it the same thing as Dayton?) and change of output inductor coils to Wurth 4.7uH, caps to 0.47uF and addition of zobel (0.47uF + 15Ohm) as suggested by Sendler. I have also replaced the each of the power caps to ELNA SILMIC II 470uF. I am driving it from Meanwell S145-24 turned up to 29VDC.

I must say the sound is very good in mid and highs after the modifications and to be fair, the bass is not too bad. It's just slightly lacking around 110Hz and below to my ears based on my playing around with the foobar2000's equalizer.

Things I have contemplated trying but haven't done yet are:

1. Change of the power caps to bigger ones.
2. Change of the SMPS to more powerful one and/or to the one which has higher switching frequency.
3. Change of input caps to bigger ones, say 10uF ... I did go from 2.2uF to 4.7uF previously.

I just wish there is a simple definite solution to this issue without having to try this and that. Or is it that there is nothing else we can do because of how things are internally in the chips?

[ZFRaudio]

"swkbkk", thank you for the detailed response.

I usually play dub, reggae and drum and bass. A significant portion of the audio information is within the bass notes (and energy for that matter) and with the sub 100hz lacking it isn't quite the same!

According to the datasheets the chips should be absolutely fine, whether its playing pink noise or 40hz sinewaves - they seem to be very capable. However, as ever, its the supporting circuitry that tends to limit these chips. As the Sure chip is made at a "bargain basement" price, then its my suspicions that some of the surrounding components are responsible!


There have got to be some more informed chaps (re. Class D designers!) out there who have dealt with this problem before who can offer an insight..

I am thinking to perhaps change the input capacitors, as these seem to act as highpasses - I am not sure of the maths, but could changing these help?

Thanks again

Zack


Ps. I run my Sure amplifier off a 26v SLA battery array with very large reserves of power

[nigelwright7557]

I use a 10uf on the input to my irs2092 amp.
The input resistor is only 3k3 which is very low for an input impedance hence the 10uf capacitor.

[teamacc]

Upgrading the input cap is not going to help your problem, as a 10uF cap will shift the cutoff frequency from +-5hz to +-3hz, which is inaudible, even with the best ears.

It hast been suggested however, that the output chips (sta5xx) have such an iron grip on the speakers that they tend to produce less bass.
I dont know why, but this seems to be the issue here.

[swkbkk]

Originally Posted by ZFRaudio

I am thinking to perhaps change the input capacitors, as these seem to act as highpasses - I am not sure of the maths, but could changing these help?

As pointed out by "teamacc" above, I also doubt if changing the input caps to 10uF from 4.7uF would improve the bass. You can also find a post by dr_vega related to this matter here Sure Electronics New Tripath Board tc2000+tp2050

The formula for determining the input low frequency pole can be found in the TK2050 datasheet available here TK2050 Datasheet pdf - Stereo 50W Amplifier Chipset - Tripath

I will be very happy to hear about a solution to this issue, but I have a bad feeling I might have to stick to the software equalization.

[kristleifur]

Originally Posted by teamacc

It hast been suggested however, that the output chips (sta5xx) have such an iron grip on the speakers that they tend to produce less bass.
I dont know why, but this seems to be the issue here.

The 41Hz implementations Amp4 and Amp11 using the same chip don't have any problems in the bass department! Quite the opposite - they have a very strong low end. It's the implementation. I'd try another manufacturer / design.

[Malefoda]

Wonder if that was only on the first short boards, wich has a different layout of decoupling caps at TP2050 ps lines?