The true experimental tweak thread

[Cornan]

29 minutes ago, sandyk said:

Others have already stated their preference for the more tonally balanced sound when the input capacitor of the LT3045 is either replaced by a larger value electrolytic capacitor, or better still perhaps, in parallel with the existing input capacitor or e.g. a 100nF replacing the original ceramic input capacitor when used with Audio .

 

I actually learned by Alex at ALT Hi-Fi that those 47uF caps on LT3045 are not ceramic ones. They are tantalum caps. Sometimes those caps needs to be tantalums as the designer intended. Not saying that anyone shouldn't change them out, but I have decided not to.

 

https://en.m.wikipedia.org/wiki/Tantalum_capacitor

[sandyk]

Just now, Cornan said:

 

I actually learned by Alex at ALT Hi-Fi that those 47uF caps are not ceramic ones. They are tantalum caps. Sometimes those caps needs to be tantalums as the designer intended. Not saying that anyone shouldn't change them out, but I have decided not to.

 

 I have previously suggested just using larger value electros in parallel with the existing input capacitors.

Tantalums are being used because you can't get ceramic capacitors in that large value, or they would be too big physically.

I try to avoid using tantalum capacitors wherever possible, due to them going s/c on several occasions in my own gear , despite being well within ratings. More recent tantalum capacitors are however likely to be more reliable.

[Cornan]

14 minutes ago, sandyk said:

 

 I have previously suggested just using larger value electros in parallel with the existing input capacitors.

Tantalums are being used because you can't get ceramic capacitors in that large value, or they would be too big physically.

I try to avoid using tantalum capacitors wherever possible, due to them going s/c on several occasions in my own gear , despite being well within ratings. More recent tantalum capacitors are however likely to be more reliable.

 

What I like when reading about tantalum caps is their low leakage properties. Since blocking leakage is a big part of my own goal when using the voltage regulator a in the DC cable path I think I will only change them out for something with better leakage properties than the tantalum caps.

[sandyk]

6 minutes ago, Cornan said:

What I like when reading about tantalum caps is their low leakage properties. Since blocking leakage is a big part of my own goal when using the voltage regulator a in the DC cable path I think I will only change them out for something with better leakage properties than the tantalum caps.

 

 You appear to be a little confused here. The leakage is like having a high value resistor in parallel with them, which is only of concern if they are being used in series between stages, such as coupling capacitors between sections of an amplifier, or at the output of a device such as  a DAC or STB etc. 

[Cornan]

1 minute ago, sandyk said:

 

 You appear to be a little confused here. The leakage is like having a high value resistor in parallel with them, which is only of concern if they are being used in series between stages, such as coupling capacitors between sections of an amplifier. 

 

So do you mean that using the LT3045s in series as I do is not the same thing as between stages? I am interested in anything that blocks/reduces leakages and do not add too much leakage of their own.

Will using the LT3045s on each side of the DC cable instead of directly after each other at the powered device end increase/decrease leakage or is it the same thing?

[sandyk]

Just now, Cornan said:

 

So do you mean that using the LT3045s in series as I do is not the same thing as between stages? I am interested in anything that blocks/reduces leakages and do not add too much leakage of their own.

Will using the LT3045s on each side of the DC cable instead of directly after each other at the powered device end increase/decrease leakage or is it the same thing?

 

Leakage, like you are concerned mainly about, is generally referred to as artifacts (SMPS ripple and HF harmonics , LF such as mains related harmonics) getting through to the output. It is the function of the voltage regulators to eliminate most of this, although common mode stuff  is a different matter as John Swenson has already discussed in detail.

Even with slower voltage regulators,(LM317 etc.)  using a series choke of a much higher inductance than described in the LT3045 Application notes will help to markedly reduce the amount of very high frequency rubbish getting through to the output via the voltage regulator .

The low ESR tantalum, or ceramic capacitor at the input of the LT3045 also helps to prevent very high frequency rubbish from passing through the LT 3045 due to their very low impedance at such high frequencies, where the LT3045 is less effective,

despite having a much wider bandwidth than  earlier voltage regulators. 

[mozes]

I got my LT3045 and tried 1 5v between the LPS-1 and my switch. If there is an improvement, it is very subtle so not sure about it. It maybe my placebo. I felt more warmth overall

[Cornan]

2 minutes ago, mozes said:

I got my LT3045 and tried 1 5v between the LPS-1 and my switch. If there is an improvement, it is very subtle so not sure about it. It maybe my placebo. I felt more warmth overall

 

Great Moussa! Did you try a single LT3045 and not two in series? I would say that the LT3045 gives a warmer sound, so I it doesn't sound like a placebo to me! ?

[mozes]

Only one, the other ones I have are 7 and 9v so won’t work. I ordered two 6v ones so I can try them in series.

[Cornan]

1 minute ago, mozes said:

Only one, the other ones I have are 7 and 9v so won’t work. I ordered two 6v ones so I can try them in series.

 

I have so far just tried them in series. Now when I finally have the whole DC chain with totally six LT3045s is sounds really great. I will try them in pairs on each side of the DC cables instead of directly in series, but since my Gopherts have banana plugs I'll need to order inline banana sockets to make it work at the Gophert end. 

Here is to give you an idea how: Gophert 6,7v>LT3045 6v>Ghent JSSG DC cable>LT3045 5v>Aqvox switch.

I have just ordered the inline banana sockets from China, so it will take a couple of weeks until I can try it out though.

[mozes]

22 minutes ago, Cornan said:

 

I have so far just tried them in series. Now when I finally have the whole DC chain with totally six LT3045s is sounds really great. I will try them in pairs on each side of the DC cables instead of directly in series, but since my Gopherts have banana plugs I'll need to order inline banana sockets to make it work at the Gophert end. 

Here is to give you an idea how: Gophert 6,7v>LT3045 6v>Ghent JSSG DC cable>LT3045 5v>Aqvox switch.

I have just ordered the inline banana sockets from China, so it will take a couple of weeks until I can try it out though.

I have a second 5V LT3045 1amp board. I am thinking of experimenting with it by using two in parallel eventhough the LPS-1 puts out only 1.1 amps. 

[mozes]

So is there a final verdict on the caps now? Should we upgrade the cermaic ones to electrolytic with bigger capacity? I have many electrolytic caps in my boxes (Mundorf and Elna)

[sandyk]

40 minutes ago, mozes said:

So is there a final verdict on the caps now? Should we upgrade the cermaic ones to electrolytic with bigger capacity? I have many electrolytic caps in my boxes (Mundorf and Elna)

 

In parallel with the existing 47uF tantalum capacitor is likely to be better.

[sandyk]

Something I intend playing around with in coming weeks.

It could be used with a 7.2V battery or a voltage regulator to supply 7.2V to a Regen etc.

Alex

 

 

Click on the image for a larger image.

[mozes]

Upon further listening to Tidal, the LT3045 is staying for sure. Tidal never sounded so good. Music is more relaxed, less harshness and lower noise floor with warmer tone that doesn’t hide details.

Whoever suggested the use of LT3045, big thank you!

[mozes]

1 hour ago, sandyk said:

 

In parallel with the existing 47uF tantalum capacitor is likely to be better.

Thanks, will try it tomorrow and report back

[Cornan]

10 hours ago, mozes said:

Upon further listening to Tidal, the LT3045 is staying for sure. Tidal never sounded so good. Music is more relaxed, less harshness and lower noise floor with warmer tone that doesn’t hide details.

Whoever suggested the use of LT3045, big thank you!

 

Great to hear Moussa! ?

That thanks goes to @tapatrick for telling us on this thread and RB2013 that found out about it.

I love them as well! ?

[Cornan]

10 hours ago, sandyk said:

Something I intend playing around with in coming weeks.

It could be used with a 7.2V battery or a voltage regulator to supply 7.2V to a Regen etc.

Alex

 

 

Click on the image for a larger image.

 

Looks interesting Alex! ? I can tell it is a supercap construction, but do you mind telling some more about it and how you intend to use it?

[tapatrick]

22 minutes ago, Cornan said:

 

Great to hear Moussa! ?

That thanks goes to @tapatrick for telling us on this thread and RB2013 that found out about it.

I love them as well! ?

yes they are a great addition esp in series. Big thanks to RB2013 Tubelover2 if you are listening...

I'm tweaking them now with various input caps and ferrite beads.. thanks to @TubeMan and @sandyk input

we are a crazy bunch indeed but it's worth it when new levels are found. 

[mozes]

1 hour ago, tapatrick said:

yes they are a great addition esp in series. Big thanks to RB2013 Tubelover2 if you are listening...

I'm tweaking them now with various input caps and ferrite beads.. thanks to @TubeMan and @sandyk input

we are a crazy bunch indeed but it's worth it when new levels are found. 

And of course @Cornan the orchestrator of this thread!

[tapatrick]

22 minutes ago, mozes said:

And of course @Cornan the orchestrator of this thread!

of course @Cornan the Great!

 

[sandyk]

8 hours ago, Cornan said:

Looks interesting Alex! ? I can tell it is a supercap construction, but do you mind telling some more about it and how you intend to use it?

 

 I intend comparing it with my recent 12V 15,000mAH battery into a JLH with around 4 FARAD simulated capacitance when powering a USB Regen, which was clearly better than a 9V Voltage Regulator followed by a JLH PSU add-on, as well as directly after a friend's HD Plex to see if it gives him a further improvement.

When using it with a 7.2V Li Ion battery, it will, shortly after the switch on delay, put around 33 FARADS of capacitance directly across the 7.2V battery, or voltage regulator set for +7.2V powered via a 12V Li Ion battery or Linear PSU.

As you can see , the overall cost will be very modest, and a good introduction to Supercap technology for me.

[Cornan]

7 minutes ago, sandyk said:

 

 I intend comparing it with my recent 12V 15,000mAH battery into a JLH with around 4 FARAD simulate capacitance when powering a USB Regen, which was clearly better than a 9V Voltage Regulator followed by a JLH PSU add-on, as well as directly after a friend's HD Plex to see if it gives him a further improvement.

When using it with the Li Ion battery, it will, shortly after the switch on delay, put around 33 FARADS of capacitance directly across the 12V Li Ion battery.

As you can see , the overall cost will be very modest, and a good introduction to Supercap technology for me.

 

I would be very, not to say extremely, interested to hear your impressions with this thing! ? If you do not mind please also share some pictures of the final design when ready. Battery powered supercaps/ultracaps is something that I have been interested in for quite some time, but never fully understood how to realize. 

[sandyk]

51 minutes ago, Cornan said:

Battery powered supercaps/ultracaps is something that I have been interested in for quite some time, but never fully understood how to realize. 

 

 What concerns me at this stage, is how well they will share the total voltage .

I will need to do further research in this area.

 

Hmmm. A bit more complicated than I first thought.

 Let's hope that the Ultracaps came from the same batch and are reasonably well matched !

[Cornan]

18 minutes ago, sandyk said:

 

 What concerns me at this stage, is how well they will share the total voltage .

I will need to do further research in this area.

 

Very inspiring experiments non-the-less! Please do share your findings. I can see greatness in it if it you get it to work properly. 100% something I would be interested in, especially for my Aqvox switch.