[justubes]

Hi Morten,

Impedance setting may well be one very and perhaps underused and not well understood by us users.

I have previously mention once I felt I had the right impendace, an adjustment just by 1 changed the sound by a large degree.

I think now with more understanding of it with different sources and music, I will be using it much more.

Morten,

I have mine set at 43, which has the best sound.Once I go any lower the sound mellows out too much.

The interesting thing also is that go up by 1, makes the sound more brilliant, tight and harder sounding. This also make the sound louder as compared with the lower setting just by -1, this is at 42. But if I go -2 (41)instead, it is still softer than impedanceat  43, but not as soft as at 42.

Is this normal?

The sweetspot of 43, though softer and I need to compensate volume up by around +2, but at this setting, all digital hardness and glare vanishes! This alone is priceless.

Is this normal?

Please remove the post in the wrong thread.

[justubes]

Morten,

To add on...

Does the quality of power supply affect the calibration?

I have 2 settings of the same impedence i.e 42, one done some time ago with a battery and another by low noise power supply.

They actually distinctly sound different!

I am not sure what is happening here.

[tortugaranger]

The interesting thing also is that go up by 1, makes the sound more brilliant, tight and harder sounding. This also make the sound louder as compared with the lower setting just by -1, this is at 42. But if I go -2 (41)instead, it is still softer than impedanceat  43, but not as soft as at 42.

Is this normal?

The sweetspot of 43, though softer and I need to compensate volume up by around +2, but at this setting, all digital hardness and glare vanishes! This alone is priceless.

Is this normal?

The question of what is "normal" when it comes to adjustable impedance is an interesting one. To begin with adjustable impedance isn't normal. I know of no other preamp that has this feature. And no doubt many designers would argue that it's not necessary or beneficial. We implemented it because our preamp design is software driven and therefore it's possible to do this. Like the mountain, we climbed it because it's there....and we could. As people play around with this feature some find no benefit while others are able to find an optimal setting that clearly improve the sound. I would not have guessed that minor adjustments would make much difference.

[tortugaranger]

Does the quality of power supply affect the calibration?

I have 2 settings of the same impedence i.e 42, one done some time ago with a battery and another by low noise power supply.

They actually distinctly sound different!

I am not sure what is happening here.

Yes. In fact power supply is most important during calibration and arguably much less so during normal operation. During calibration, the preamp is actually a closed loop measurement system where DACs generate voltages that are passed through LDRs with the results measured by ADCs.

That said, it's not immediately obvious to me why the results would sound distinctly different. It could be that one calibration data set is a 1 or 2 dB louder than the other and so the perception is a difference which may only be a matter of volume. Then again, there are numerous variables working here so I wouldn't expect identical results.

[justubes]

Does this point to impendenca has an effect on volume. I find that lower impendence to sound softer mellow, yet more natural and relaxing.

I really dont understand impendence, but is does alter the electrical voltages and thus affects drive.

This is intriguing to the possibilities of further tuning.

We discussed previously on low noise psu in which I had used a tps74, I had switch this out for another low noise reg. I recall I possibly did a recal using batteries for that 1 setting, thus with a different regulator.

While music is on and toggling the impedence setting on the fly, I clearly hear the differences.

A friend was baffled how I could change the sound, it sounds as if I had change a different power cord all together.

If I go down another point which used the old regulator previously doing the autocal, there is a resemblance in the sonics percular to the autocal done then, I am almost afraid to repeat the autocal in fear I may lose the familar sound I was getting.

[craig sawyers]

Does this point to impendenca has an effect on volume. I find that lower impendence to sound softer mellow, yet more natural and relaxing.

I'm baffled how such small changes in the resistance setting of the Tortuga results in such noticeable changes in sound quality.  Now I can hear the difference between 20k, 50k and 100k, but that is probably more to do with the output resistance of the Tortuga producing additional attenuation when loaded by the power amp input resistance.  Suppose you listen at 50 on the display, with a nominal attenuation of 14.32dB.  Suppose for the moment that the input resistance of the power amp is 100k (many are lower than this).

At 20k, the amp loading reduces the volume (as compared with no loading) by 0.27dB.  At 50k by 0.65dB and at 100k by 1.26dB.  So you have to be very careful how you interpret what you are hearing when you switch the Tortuga between resistance settings on the fly.  The odd dB shift in volume is clearly audible, and easily interpreted as it sounding softer.  If you can hear the effect of changing in 1k increments something odd is happening.

Have you done a blind trial with other listeners?  Do they hear the same thing as you?

[TJHUB]

It is my understanding that the adjustable impedance of the Tortuga affects the source, not the output to the amp.  The higher the setting, the less stress put on the output of the source (most cases a DAC).  Morten can correct me if I'm wrong.

My DAC seems to like a very low setting.  20k is good, but 10k and even 5k produce better transparency and dynamics.  As I increase the impedance to 40k, 60k, 80k, and 99k, the sound softens, loses dynamics, and even colapses the sound stage.  99k is not listenable for me.

In all of my testing, I never heard significant changes with small impedance changes.  But every source can behave differently.  I do know consistency is key.  If you run a calibration sequence with different power supplies, you can expect more volume changes between calibrations.  I would rerun all calibrations that are not done with the same power supply.

I hope I'm correct with my statements.  If not, then I'll need a refresher in how this works.

[justubes]

In my setup, 1-2 steps ether up or down is clearly audible. The lower the setting at the sweetspot makes a nice and noticeable difference.

At other settings outside this sweetspot, it was not really audible, nor would i bother.

I know for a fact the autocal on a different psu seems to contribute this this effect.

Presently, i am delighted with the sonics now, so do not want to recal until it is really needed and as there isnt enough spare slot to auto the new psu.

I dont want to lose the old setting either, i tried 1 recal and it is indeed different sounding.

The db difference is also noticeable, but more so when you have the sweetspot setting, the wrong setting doesnt nearly sound as good and difficult to distinguish from slight adjustments.

[craig sawyers]

Morten - here's a thought for those who know the input resistance of the amplifier that follows the Tortuga.  In the new design, if there was an option to input that resistance, then loading-induced loudness changes could be accurately compensated for by a change in attenuation.

In that way decoupling loudness changes from sonic changes could be done.

[33na3rd]

It is my understanding that the adjustable impedance of the Tortuga affects the source, not the output to the amp.  The higher the setting, the less stress put on the output of the source (most cases a DAC).  Morten can correct me if I'm wrong.

I had assumed that changing the input impedance would automatically change the output impedance too. This is very interesting!

[craig sawyers]

Well yes it does.  But the effective output resistance is the actual output resistance of the Tortuga in parallel with the input resistance of the power amp (or whatever) it is connected to.  Which changes the attenuation.

[tortugaranger]

I had assumed that changing the input impedance would automatically change the output impedance too. This is very interesting!

Yes and no. Input is nearly fixed for any given impedance level whereas output impedance varies as you change volume (attenuation).

Input Impedance

Whether it's a potentiometer or a pair of series/shunt LDRs, what you have is a voltage divider.  With pots the input impedance is always fixed. It's the rated resistance of the pot from input to ground which per the sketch below is Ra+Rb. Ground is the bottom horizontal wire and the wire coming out between Ra and Rb is the pots wiper ...OR.. the output of the voltage divider (same thing).

With LDRs is can also be fixed but not necessarily. It depends on how you control the LDRs. In the sketch the series LDR is Ra and the shunt is Rb. Again, it's the sum of Ra and Rb that determines the input impedance. Unlike with a pot where this sum is fixed, with LDRs you must control each of these independently. Which we do in our LDR preamps.

The way a voltage divider works is per this formula:    Vout = Vin x Rb/(Ra + Rb) or Vout/Vin = Rb/(Ra + Rb). The resulting attenuation expressed in decibels is dB = 20*log(Vout/Vin)

What we do in our LDR preamps where you can change the input impedance is we compute the combinations of Ra and Rb that give us the 70 attenuation steps we need while trying to maintain the sum of Ra + Rb equal to the target input impedance. In practice this is not possible while also attenuating down to -60 dB so what happens is the impedance has to go up from the target as volume gets turned down low (usually around step 20 and down).

Output Impedance

Output impedance is different. It's not fixed. Not in a pot and not with LDRs either. The way to look at output impedance is just flip it around and look at it as though it was just another input impedance with the output jumpered. That's illustrated below.

Unlike with input impedance where Ra and Rb were in series, here Ra and Rb are in parallel. With parallel resistors(impedances) the formula is R = Ra x Rb/(Ra + Rb)




Tying all of this together you get something like the graphic below. This shows an attenuator with 20k input impdance - the green line. Note that it's constant until you get down to around -50 dB and then it increases in order for the attenuation to continue down to -60 dB. It's just how the math works.

Now note the output impedance which is the yellow line. As volume increases it rises from a very low value (100 ohms) and then peaks at -6 dB (loud! ) and abruptly falls back down to 100 ohms as attenuation approaches 0. This is how every pot works in every preamp ever.



Of course all the above is simply looking at the attenuator in isolation and not hooked up to anything. When you attach an amp to the output of the attenuator you add an additional resistance in parallel. So you have to do the R = Ratten x Ramp / (Ratten + Ramp) calc to get the effective impedance as seen by your source (DAC, phono preamp etc. )

[JosepHLG]

I recently just installed my new Tortuga attenuator. I find that for my system the best impedance setting for my balanced unit is about 30. Lower or higher settings result in unbalanced stereo channel volumes requiring significant channel balance adjustment. In fact, before I adjusted the impedance to 30 most of the sound was coming from only one speaker and I thought my system had a broken component. When I tried different impedance values from minimum to maximum the center focus was only correct at around 30 to 35 where the volume of the two channels is about equal. This is also where the sound is most open and dynamic and neither too thick nor too lean. This surprised me because the Goldpoint balanced attenuator I replaced with the Tortuga unit had a fixed impedance resistance of 10k and had only a slight mismatch of volume between stereo channels. Fortunately, the Tortuga's impedance adjustability allows it to perform optimally in my system. At setting 30 it sounds superb.

I am still experimenting with the impedance adjustment to

[JosepHLG]

To complete the unfinished sentence in my last message, "I'm still experimenting with the impedance adjustment to" determine if a value slightly under 30 or slightly over 30 results in the best sound. Intuitively, a digit or two either direction should not make an audible difference, but posts by others say they can hear differences by changing the impedance value by only one unit.

[TJHUB]

There should be no balance issues.  Are you running auto calibration after each change to the impedance?   If not, that's the issue.

[JosepHLG]

After each impedance assignment I ran Auto Cal. I did this separately for each of the following assignments: 1-10k, 2-20k, 3-30k, 4-40k and 5-50k. The last Auto Cal followed the 50k assignment to setting 5. When listening and changing the settings from 1 to 5 (10k to 50k), only 30k has the correct balance. Maybe I'm not doing Auto Cal correctly. When I change the impedance for a particular setting I then turn off the preamp, turn off the source, and then press the center button on the preamp remote to initiate Auto Cal and let it do its thing until the unit shuts itself off automatically.

[tortugaranger]

After each impedance assignment I ran Auto Cal. I did this separately for each of the following assignments: 1-10k, 2-20k, 3-30k, 4-40k and 5-50k. The last Auto Cal followed the 50k assignment to setting 5. When listening and changing the settings from 1 to 5 (10k to 50k), only 30k has the correct balance. Maybe I'm not doing Auto Cal correctly. When I change the impedance for a particular setting I then turn off the preamp, turn off the source, and then press the center button on the preamp remote to initiate Auto Cal and let it do its thing until the unit shuts itself off automatically.

There shouldn't be noticeable changes in channel balance between impedance settings. I've not observed this behavior before nor has it been mentioned prior to this instance. It makes my wonder if there's some kind of interaction between the preamp and the either the source's balanced output stage or the amp's balanced input stage such that changing the impedance somehow skews the channel balance. If so then this may have more to do with the design of the respective balanced input/output stages rather than a true left/right channel balance with the preamp. Some balanced input/output designs are less robust than others and can be influenced by external impedances/loads.  Not a very satisfying response but at least you were able to find a sweet spot where it all works and sounds best.

[kernelbob]

You might try swapping your source inputs at the Tortuga (left/right) while at an impedance setting where you hear an imbalance.  If the imbalance then switches to the other channel, that would seem to indicate that one of the source's channels doesn't like that input impedance.  Similarly, try that same switching of the outputs (left/right) and see if the imbalance moves.  Finally, just to rule out a synergy (or is it anti-synergy?) between the source and amp at a particular input impedance.

One other thing, do you hear different degrees of imbalance at different attenuation settings?

Best,
Robert

[justubes]

Just to chime in, i have been struggling with this have encountered the same issue and need to adjust the balance +2  to get the sound from both channels right and centered as one side is louder on my balance unit.

It has been confirmed that on another impedence setting which i have save, it is yet again different and requires a +3.

I have had an issue getting autocal done which i don't know if it is related, it worked previously but keeps hanging, problem is i only have 1 display. I am afraid to touch it as it is as it has not affected my enjoyment from my system and not wanting to do another autocal in fear this my make the problem worst.

One of my impedence settings is even worst, extremely soft on one side.

[tortugaranger]

Just to chime in, i have been struggling with this have encountered the same issue and need to adjust the balance +2  to get the sound from both channels right and centered as one side is louder on my balance unit.

It has been confirmed that on another impedence setting which i have save, it is yet again different and requires a +3.

I have had an issue getting autocal done which i don't know if it is related, it worked previously but keeps hanging, problem is i only have 1 display. I am afraid to touch it as it is as it has not affected my enjoyment from my system and not wanting to do another autocal in fear this my make the problem worst.

One of my impedence settings is even worst, extremely soft on one side.

These symptoms suggest at least one LDR may be out of spec and not able to calibrate properly. Channel balance shouldn't change noticeably with impedance setting. The calibration algorithm works exactly the same regardless of the impedance setting.