Hi All.

I am Justblair, and I would like to apologise for the teeny weeny schematics and thank you for mentioning it.

One reason you might wish to play with the input resistors would be if you were having issues with the input impedance of your amp vs the output impedance of your source.

More likely would be if your source is not giving either too high or not high enough voltage at the output. Some have modified the input resitors to give higher gains for using portable devices with their amps.

I have added a closeup of the tripath circuit for just now until I get my site to magnify images on click as it is supposed to be doing!

Thankyou for linking to my site, I want it to be as useful a resource as possible to people like me.



Blair

Thanks for the warm welcome.


I did indeed update the page. Its a temporary solution till I work out something more elegant. I hope it's not shameless self promotion to point out I have a list of articles based on the Sure Electronics TA2024 amplifier. Maybee not what you are discussing here, but as the circuits are so similar it may be of interest.

Blair

I have been having one or two issues with images on the site. That one is one I have to get round to sorting...

Blair

Image rectified. Its a Gif thing obviously.

I haven't been able to find those capacitors looking at the pictures, but I have not manageed through the 70 pages of forum here. can someone post a decent picture of them?

The LPF capactitors that I removed from the sure board were 100nf values. If yours are 100pf, then the -3db point of the filter will be much much higher higher have you had them out to measure that value?

I would be doubtfull if 100pf is a right that removing them would make any audible difference.

One thought. If you take a resistance meter, and can find that their poles are attached to both the signal in and ground, it will be likely that they are serving the same function as on the sure boards.

Oh and another thing. There is a thread going on at DIYAudio.com

here

Its about the sure boards (you will find that my information on what modifications to do came from there) but about halfway through Arjen introduces the same boards that you are working on. I have not been reading all of it, but I would be suprised if some of the mods you are doing are not urning up there as well...

Ah thanks for the schematic. Yes indeed the caps you are talking about are part of a LPF. The designer has used a second order HPF alongside the two inductors L1 and L2.

The reasons for doing so are basicly to make the amp measure better for noise. However there is a difference between measured noise and perceptable noise. I would remove both from the signal path.

I would not rush to hard. The very small value of the caps means the LPF will be at a high frequency. You may not notice any difference. The schematic had inductors in the signal line.

This means a filter that's designed to work at a high frequency (out of human hearing ranges). Using the inductor and the cap means that the filters slope is steeper.

The sure boards that I have worked on used a very much higher value of cap 100nf. So the LPF cuts in at lower hz, hence dulls the top end.

Of course if I am reading your posts correctly, the boards you have differ from the schematic in that the inductor is not there. It is very possible that you will find they have used higher values to compensate for the lack of inductor.

Blair

Hello. I hope that I have not upset anyone by cross linking the two threads. I did do with the best of intentions. If I have done wrong, please accept my apologies.

As for asking Jazzy if he would like to contribute to my site. I make an open invitation to anyone who has a project or anything audio orelectronics that you wish to share to get in touch. My aim is to build a site that anyone like us who is interested in the hobby can find inspiration and knowledge that will further their pastime. The problem with threads are that a lot of the knowledge in forums becomes diluted amongst so many posts. To distil this into a PDF or a web article is easier for the beginner.

The work that you have all done here is of interest to others, particularly the members at DIYaudio. They have just discovered the amp modules you have been playing with up till now. I think personally it is brilliant that knowledge can travel so quickly and that they will get the benefit of your experiances. Hopefully you may get some benefit from theirs as it builds with this product.

Humbly

JB

This post has been edited by justblair: Sep 19 2008, 03:08 PM

No offence taken. Saw you pitched in at DIYaudio. Think you may be doing a fair bit of bridging info as they get a chance to try out your modifications and come up with some of their own...

I may have to get hold of a couple of these boards so that I can join in the fun!

Definately up your voltages. I'm using an smps with an adjustable voltage and I can here a difference in the solidness of the bass. Drums have more impact at higher volts. Thats whats good in my system. I wonder wether voltage might be a tuning thing or an improvement thing?

It would be interesting to see what the cap values are for C23, C24. The sure boards have different resistor and cap values than their schematic suggests.

I have not tried out Arjens board yet. I am having too much fun with me Sure Tripath board.

On that note, just tried out another modification that may be of interest to you. I have seperated the power lines for the left and right channels. Inititail testing is very positive.

Powerline modification part 2

Good work on the manuals Jazzy!

You can try at this point to change the values in your output filter. 10uh coils are about right for 4ohm speakers, 15uh for 6ohm and 22uh for 8ohm I think. The tripath datasheet makes mention of it I think, but don't trust my memory, check before spending cash.

I dont know how much benefit there is to matching the output to the speakers impedance

When you took the caps out, did you connect signal ground to the bias pin?

Couple of points:

Mica caps are not normally used in the signal path in Audio applications. The general consensus is that polypropylene (though some favour paper in oil) caps are best suited to signal line DC blocking. Thats not to say that you shouldn't experiment, but if your paying lots of cash to try it out, you might think twice and invest this in some well reviewed polyprops.

Secondly 10uf is almost definately to high a value for the input caps. If you look at my online calculator at www.justblair.co.uk you can enter the different cap values and 10uf gives you a HPF set at 0.8 hz. But remember that is the -3db point. The whole point of the input capacitor is to block DC. At this value, you wont be doing this very cost effectively. A 4.7 uf cap gives you a 1.69hz HPF, even that is probaby excessive. 2.2uf or 3.3 would be the ideal, 4.7 at a push.

If you wish to parallel the input cap, parrallel with a very small value for the second cap. MKP1837 at 0.015uf is a decent value to choose.

Jazzy. Every capacitor has its own "signature", although some are more transparent than others. The contruction of the caps means that they do not pass frequencies above their filter point equally. That is not to say that some of the "boutique" capacitors are necessarily better performers than the cheap ones, they all just colour the sound differently.

The best solution is to have no capacitors in a circuit altogether. However the capacitors have a very useful property. They block DC signals while allowing AC signals (ie Audio) to pass. To work out why, look inside the cap. You have two layers of conductor, seperated by a very thin layer of insulation. What happens when electrons are removed from one side than the other? (ie a positive charge) is that electrons on the opposite layer are attracted to the positive charge, they flow in.

Of course when enough have flowed in, they stop, and you have no current. Ie the DC has been blocked.

Add some electrons to the now positive side, and the electrons on the other side will no longer be in balance and will flow away from the plates until equilibrium is now reached.

Remove them again and the opposite occurs.

Whats happening now is that the AC signal occurs on the other side of the cap as a reaction to the ac signal on the source side, or to put it another way, AC has been allowed to pass.

The fact that you get a filter this way is almost consequencial. But you can use this "by product" in your design. If you want an amplifier to feed a high range speaker you could use a very small input cap to restrict not just the DC but also the slow changing AC. (lower frequencies)

I cant think right now how to explain why caps filter low frequences rather than just DC. I'll try and come up with the words to express it. But basically it just does! Its to do with the charging and discharging times. Thats where the resitor comes in. It reduces the current, meaning the cap takes a finite time to discharge and recharge. If the frequency of the signal is low enough, it chaneges alplitude (voltage) more slowly. The cap discharges quicker than the change in source signal, so the source signal does not pass through. Kinda???!!!

My explanation is really bad on the filter bit... Sorry, best thing to do is just accept that it does!

Well a low pass works in a similar fasion, but this time the cap is connected to ground. As the higher frequencies pass through the cap, they are discharged to ground and removed from the signal path. The lower are sheparded on to the amplifier.

The Sure boards I have, had this arrangement in place, but unfortunately the caps they used were large enough to pass some audible frequency through as well.

Removing them returned the highest of the audio frequencies and improved the top end sound. You can also use a coil here, as inductors prevent high frequenies from passing. Thats because they use a magnetic field to filter.

If you dont have a low pass filter, your amp is reproducing the high frequenices outside of the audible frequency spectrum. (HF noise from other components for instance).

Some amps dont like doing this and it can make them oscillate, but the tripaths are stable at the higher frequencies. Of course with instrumentation you can detect the amplified high frequences (or "noise") and without the low pass filters the amps measure worse for Signal to Noise Ratio (SNR). Humans though cant hear this noise, so some audio equipment with apparently high SNR sound better when you think they should sound terribel looking at the figures.... Non-oversampling DACs for instance produce high SNR yet are loved by their owners! (ie me!)

A commercial manufacturer would on the other hand find this high SNR bad, but thats more for commercial reasons, SNR has for a long time been banded around as an indication of high quality... Its not as simple as that, the frequency of the noise is just as if not more important.

The low frequency isn't grounded, but rather it's blocked. However, if you design your filter to a low enough point, it doesn't matter.

Humans have limits to what low frequencies we can hear just as we have for the high frequencies. You set your filter low enough and we can't detect either those frequencies absense or inclusion in the signal.

On top of that, your speakers are far more of a limiting factor here. At 50hz my Mission 753's are half as loud as at 100hz, and lower than that the frequency they just get worse.

Even the largest of pipe organs only reach down to 15hz or so. If your filter is set at 5hz, you dont lose any of the music.

The maxim products use a feedback system to get rid of the DC on the input side, and an intermodulation system for the switching signal on the output.

No caps.

However that doesn't mean that they sound good using this system. Its notable that they also have an internal switch to remove the no filter trick. If you are using them like we do, ie with the speakers far away from the amp, its recommended to add an external output filter. They also point out that you get high SNR with no output filter....

I have a couple of the new MAX9744 chips in my parts box. I was thinking about making an amp with them at some point...

**edit Beaten to the punch****

This post has been edited by justblair: Oct 4 2008, 01:29 AM

I thought you may be interested, A chap called Ales has submitted an article to www.justblair.co.uk about the boards that you have been working on.

I have one on its way to me so I will be posting some more articles on the amp module in the future.

The article is here....Ales article on the TA2024 amplifier module

Hope its of interest...