Split Topic: Impedance mismatch...

Over the last weekend I took my Dac's, both TwinDac + and Crazy T to an audiofile friend of mine. His system is largely DIY and includes wooden cone speakers and monstrous amps based on GM70 lamps that heats the entire room. The system sounded wonderful, very involving and with good image and scene. We put the DAC's on test and were very disappointed with the sound. I couldn't believe my ears especially due to the fact that his source was CEC TL2 going into Sonic Frontiers DAC. Good components, yes but nothing more. Since I left my brain keeps coming back to it and searches for reasons. Impedance....
I like the way my system sounds and at least this is a relief!

Also, to my big regret I finally found out that I ripped all my CD's not optimally and have to redo it... and got to search the net for clues on how to do it right. but this is off-topic.

locked

4918 views

well, IMHO 2.8k (if i'm right here) impedance of mirko's DACs really limits it only either using them in the systems with an active or a bufferred passive preamp or with a power amps with very high impedance, say 100k. which is i think what mirko has at home as he is a tube guy.

i guess this is what you've heard there.... an impedance mismatch with your DACs.

locked

4543 views

Wait a sec, Django TVC (same transformers as Crazy) and Crazy that Bert uses are not buffered are they? And yet they work well together. Could you clear it out a bit, please

locked

4526 views

well i don't know how good they work, it only depends on what load they feed (the input impedance of the power amps).

in my case, 2.5k feeding 10k solid state amp via TVC didn't produce good result and it sounded much better via buffered LDR attenuator with 3ohm output impedance. than feeding 100k tube amp showed much much better sound.

i don't know what's the input impedance of crazy A but i expect it to be very high in order to accomodate such a high imp source as different mirko's DACs (all 2.5k i think)

bert will probably chime in...

EDIT: mystery solved - from a twindac+ site:

Analog Output:
Output impedance: 2.5 kohm
Maximum output voltage: 0.8V peak
Minimum load: ≥ 50 kohm

So it's just as i've said above

locked

4430 views

Thank you!

Is your current system in Bert's Showroom by any chance? I remember you were experimenting with the Swings and thought that Crazy A's did their best in controlling the Swings... I guess you are happy with the Swings but are using some interesting amps...

Dmitry.

locked

4416 views

Good that you've reminded me about that! I should provide Bert with photos and some nice story.

Crazy A is nice but I'm a sucker for SET tube amps so I got custom 10Y SET monoblocks built in order to get the Swings singing. I personaly like how it sounds, plenty of emotions, good tonal colours and amazing transparency thanks to the direct coupling (pure Loftin White circuit) between the stages (no coupling caps nor interstage transformers employed), while a good grip thanks to a generous power supply. Less coloration than usual 300b business thanks to the amazing 10Y tube.

The system now looks like this:

speakers: Oris Swing
power amps: Tatic Audio 10Y SET monoblocks 2x4W (A2 class)
preamps: Bent TAP-X or Myth LDR
turntables: Stabi S/Stogi S/Benz Micro MC Gold or Stabi 2/Stogi/Denon 103
phono preamps: Pass Pearl/Xono or EAR 834P (for sale) or Zoran Dimovski RIAA
step up transformers: Denon AU-320 or Haufe T-7885
CD transport: Philips CD-160 (modified)
DAC: TwinDAC+ (for sale) or AltmaGC Attraction (for sale), with UPCI gizmo
tuner: Cambridge Audio C-50
cables: Duelund speaker, Hitachi LC-OFC ICs.

locked

4501 views

Hi Dmitry,

The CrazyA's have +100k input impedance and the CrazyV is less "transforming" impedances making it a rather good passive combination with full bandwidth in that setup (CrazyT > CrazyV > CrazyA) and short interconnects.

I once measured the output on my amps using the TD+ and a Django and compared that to the CrazyV...with the Django the high frequencies already were -3dB at 20kHz (limiting bandwidth) where the CrazyV is still at 0db at that frequency.

Using a low output impedance device and this bandwidth limitation did not occur...

This directly results in less quality bass too (more wooly, less dynamical and less "drive"), much less a problem with the CrazyV...

I am not much of a real technician to go much deeper in theory but I guess this explains a bit anyhow...

Bert

--
BD-Design - Only the Best!

locked

4532 views

Hi Bert! Long time

Using a low output impedance device and this bandwidth limitation did not occur...

What did you mean by this ?

By the way, you were right warning people to rip things properly. I thought I did. Well, I didn't

Dmitry.

locked

4436 views

Using a low output impedance device and this bandwidth limitation did not occur...

What did you mean by this ?

he meant using a low output imp source drives the signal through jango without any roll-off - just what i wrote above. too high opt impedance of a source is usually detrimental unless the rest of the system is not somehow tuned to it, just as it is the case of crazy setup.

locked

4493 views

I am definitely learning here

Any advice how I can tune Django to produce lower impedance? I understand that you need some gizmo but what exactly, I am not a very tech guy as you can see.
Django is now feeding Border Patrol 300B, an amp that Bert used to use before he ventured into his wonderful Crazy line.

Thanks !!

locked

4502 views

Hi Dmitry,

Any advice how I can tune Django to produce lower impedance? I understand that you need some gizmo but what exactly, I am not a very tech guy as you can see.

The only thing inside the Django is the transformer itself and you can't change anything in there. Which type of transformers does your Django use?

Django is now feeding Border Patrol 300B, an amp that Bert used to use before he ventured into his wonderful Crazy line.

Nice amp indeed with high enough input impedance...

Things to improve "drive" (if that is the problem) can be done with short and high gauge interconnects using good plugs (low gauge makes the sound smaller and more lean in my experience...).

Passive chains are rather sensitive to the used interconnects, experimenting here would be worthwhile...

Bert

--
BD-Design - Only the Best!

locked

4438 views

Hi Bert!

The only thing inside the Django is the transformer itself and you can't change anything in there. Which type of transformers does your Django use?

I believe these are the same transformers as in Crazy V, from Thailand.

Things to improve "drive" (if that is the problem) can be done with short and high gauge interconnects using good plugs (low gauge makes the sound smaller and more lean in my experience...).

Passive chains are rather sensitive to the used interconnects, experimenting here would be worthwhile...[/color]

I must be silly, I don't understand what low / high gauge interconnects and plugs are ... I should but I don't :blush:

locked

4470 views

Hi Dimitry

I believe these are the same transformers as in Crazy V, from Thailand.

Then this part seems okay. All switches at the back not set to "float"?

Earth should be connected on both ends, the sound becomes fuzzy if you don't..

I must be silly, I don't understand what low / high gauge interconnects and plugs are ... I should but I don't

Yes you are anyhow but related to the gauge talking it migth be me being silly or just not using the right words.

Try this, thick wire (cores and ground) instead of rather thin wire and on top of that good and clean contacts... (less resistance).

Bert

--
BD-Design - Only the Best!

locked

4449 views

Bert!

Now we're talking! I finally understood. Will check the earth, will clean the contacts will not switch to thin wires (you know some people swear by it). Oh! I just realized that I might have my interconnects made of those very thin silver wires, many of them of course, intertwined. Just like these Kondo cables I think

http://www.soundscapehifi.com/kondo-ksl-cables.htm

"4 cores shielded interconnect cable 99.99%Ag 0.1mmD x 52 strands x 2 core (total 104 strands) 99.99%Ag 0.15mmD x 20 strands x 2 core (total 40 strands). All pure silver wire is long term annealed to provide mellowness for total music enjoyment. It is a combination of different diameters (0.1 and 0.15mm strands) based on extensive listening tests.

Tests have shown that the narrow gauge wires give good delicate musical performance but lacks a certain amount of body, while the converse is found to be true for the thicker wires. This "catch 22" situation is cured by the use of their litz construction where 0.1 mm wires are bundled together into 52 strand wires. However, when selecting 0.1 mm wires, they also found 0.15 mm gave good but slightly different sound. Therefore, to cover all aspects of sound quality, the new LP uses bundles of 0.1 mm and 0.15 mm wires".

locked

4415 views

I am holding the output resistors in my hand now (from the CrazyT dac unit) and the value is 2K (also I mesure them just for the case and it is 2K) :grin:

it is petty that we don't have possibility to send the photos here?

locked

4420 views

Thanks for measuring...

it is petty that we don't have possibility to send the photos here?

You can upload photo's (see upload button at the right side of the edit screen while writing your message) or link to a picture elswhere..

Bert

--
BD-Design - Only the Best!

locked

4495 views

Hallo !
I did myself a lot of measurements on impedances.
Important thing ist to measure in the frequency range of interest - I´d suggest 5 Hz - 100kHz, and to measure with an vector analyzer and/ or TDR.
My results are, that the average RCA connector ( except WBT Nextgen, which are about 75 Ohms) has 23 Ohms impedance.
High End audio cables vary from as low as 19 Ohms to 120 Ohms.
Technically that means, if you dont terminate that cables with their specific impedance, you get reflections back to the source. The higher the source resistance, so worse it gets.
These reflections means of course also variations in dynamic frequency response.
For me that is reason enough, to match everything to the impedance, given by unchangeable parameters, like interconnects. Also it means of course to use very low impedances. Anything below 50 Ohms is quite good.
To build output stages, that can drive under dynamic conditions a low impedance load is not quite easy.
The standard measurements for impedance don´t mean anything, if not paralell the distorion am intermodulation is measured and low.

Best regards
Stephan

locked

4597 views

Stephan, as usual after reading your post, I am more confused then before It took me several days to digest it and I still don't know what to think
:confused:
This doesn't stop me to recommend to others your player as the very best :clapping:

locked

4440 views

no photos then :wink:

locked

4397 views

RSS Feed of thread

Split Topic: Impedance mismatch... (BD-Design)

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...

Impedance mismatch...