[robertopisa]

Using the description here on how to choose the impedance (from 1k to 99k) of Tortuga's pre
http://www.tortugaaudio.com/ldr-passive-preamps-with-adjustable-impedance/

I prepared a simple spreadsheet that computes the resulting impedance ratio "(Tortuga+amplifier) : source", where R1 is the source impedance, R3 is the amp impedance, and R2 is the wanted Tortuga's impedance (see the yellow cells).

download file Excel

The preferred choice for R2 can be then plugged in Morten's spreadsheet

http://www.tortugaaudio.com/download/Impedance_Attenuation_Schedule_Rev_012215.xlsx

[rajacat]

The tour unit has a pair of V2 boards. I'm not going to swap them out for V2.1's but I am going to upgrade to the 2.1.x firmware and as part of that run both boards through a one-time pre-calibration sequence using precision 0.1% Dale resistors. The non-LDR hardware in each board has enough inherent spec variation such that the auto-cal results can be off for reasons unrelated to the LDRs themselves. Said differently auto-cal may think it's measuring resistance level R when it's really R times some error factor. Pre-calibration is a way of squeezing out some of that inaccuracy. When you step back from all the details, auto-cal is simply an embedded resistance measuring instrument. Pre-calibration is the calibration of that instrument. The net result is a preamp with more precise channel balance which usually translates into better sound stage and stereo imaging. It's all good!   

Morten,

My LDRx requires ~ 6 pts. balance adjustment. Does my unit need a pre-calibration or is this just the sensitivity of the LDRx adjusting for  some inherent imbalance in my system? I did reverse the channels to determine if it was a weak channel in my little digital amp but no difference was detected. I did do this with music, not test tones. Is a 6 pts. adjustment significant?
Is there a breaking period in which the LDRs' equalize?

[tortugaranger]

Morten,

My LDRx requires ~ 6 pts. balance adjustment. Does my unit need a pre-calibration or is this just the sensitivity of the LDRx adjusting for  some inherent imbalance in my system? I did reverse the channels to determine if it was a weak channel in my little digital amp but no difference was detected. I did do this with music, not test tones. Is a 6 pts. adjustment significant?
Is there a breaking period in which the LDRs' equalize?

Off hand that sounds like quite a bit. If you haven't done so recently I'd run it through auto-cal and see if that changes anything. Beyond that I'd recommend first updating to latest firmware (2.1.7) followed by rerunning auto-cal. Even without pre-cal, the 2.1.x firmware renders much better channel balance. Pre-cal is icing on the cake.

LDR's seem to either go bad early on or last a very long time. By "go bad" I mean drift beyond ability to calibrate. Our experience is less than 1% behave this way. Too early to say if that's what is happening.

[rajacat]

Off hand that sounds like quite a bit. If you haven't done so recently I'd run it through auto-cal and see if that changes anything. Beyond that I'd recommend first updating to latest firmware (2.1.7) followed by rerunning auto-cal. Even without pre-cal, the 2.1.x firmware renders much better channel balance. Pre-cal is icing on the cake.

LDR's seem to either go bad early on or last a very long time. By "go bad" I mean drift beyond ability to calibrate. Our experience is less than 1% behave this way. Too early to say if that's what is happening.

Problem solved! I ran auto-cal and the channels now balance.
I forgot to to run auto-cal after fooling around with the  impedance adjustment function. IMO the impedance matching adjustment is like the "killer app" in that makes the Tortuga adaptable to almost any setup.
 BTW the LDRx works well with the 800 ohm output impedance of my DAC and the amp's 30K input impedance. I tried raising the Tort's impedance to 99k but it was too much. I now run the Tort @ the default 20k. The volume level for listening to music is generally in the 35 pts. to 55 pts. range. My rig has never sounded better. I'm seeing deeper into the music.

[tortugaranger]

[moved this thread to its own topic]

Glad to hear everything worked out. There's no doubt that setting up a new impedance setting or changing the level of an existing setting is a bit messy. But it does work! 

[robertopisa]

Hello Morten,

can you please say again how to perform a pre-cal? I have a DIY V2 dual-mono boards but pressing the enter button (when Tortuga is off) makes the auto-cal start. Apparently, I have no option to run pre-cal.

I also have some questions that are maybe of interest for people having dual-mono balance V2 boards.

In the previous firmware, 20k impendance was such value for single ended boards, but double value (40k) for dual-mono balanced boards.

Is this the same in the new firmware? If it is so, then dual-mono balanced board has only choices for 2, 4, 6, …, 99x2 with steps of two. I was using 2.5k (which is therefore 5k) but I cannot set it in this way and I have few choices as my amp has input impedance 10k

Thanks
-Roberto

Edit. This is what I found in a previous message. Where can I get the resistor module?

Precision pre-calibration process: Using precision temporary plug-in 0.1% Vishay Dale resistor modules it’s now possible to pre-calibrate the auto-cal circuit thus substantially improving the overall accuracy of auto-calibration. This further improves channel balance with commensurate improved stereo imaging and sound stage. We now do this as part of commissioning each V2.1 board prior to running auto-cal. This process is only warranted with new boards or after making hardware changes such as changing opamps. This process is also applicable to existing V2 boards but will require updated software (free download) plus the resistor modules. Tortuga Audio will soon be offering a relatively inexpensive 4-pack of pre-calibration modules as an option when purchase the V2.1 or as a separate purchase (the DIY’er can also build these themselves).

[rajacat]

The bootloader program doesn't recognize the LDR as an USB input?  Anybody else have this problem?

[robertopisa]

It does recognize it. But if I press the enter button, then auto-cal starts. I read that I have to press up/raise three times after that to start pre-cal. I also read that I need the second display, so I will order it.

The bootloader program doesn't recognize the LDR as an USB input?  Anybody else have this problem?

[tortugaranger]

The bootloader program doesn't recognize the LDR as an USB input?  Anybody else have this problem?

The bootloader program does not recognize that it's plugged into the preamp board simply by plugging in the usb cable. You have to put the preamp into bootloader mode. The preamp only enters bootloader mode when power is applied. When you apply power to the board it starts the bootloader code  within the board which then waits for only 5 seconds. During those 5 seconds the companion bootloader program running on your PC will see the preamp and you have to press the connect button on the app to lock in that connection. If you don't, the preamp terminates the bootloader code. If you miss the 5 second window you have to pull power from the board, wait a few seconds and then plug power back in.

[tortugaranger]

It does recognize it. But if I press the enter button, then auto-cal starts. I read that I have to press up/raise three times after that to start pre-cal. I also read that I need the second display, so I will order it.

Sounds like we're confusing topics here. Auto-cal and pre-cal have nothing to do with the USB connection or uploading firmware.

I will provide additional info on pre-cal soon but I strongly caution owners to not mess with precalibration unless that are fully prepared to do so. If you fiddle with pre-cal and do it wrong there's no graceful exit and your board will probably not work right until you get through it correctly.

If you own a v2.1 board there's no need to rerun pre-cal. Owners of v2 boards can improve auto-cal accuracy by doing a one time pre-cal but pre-cal requires 2 different values of precision reference resistors be plugged into the board instead of the LDRs.

[rajacat]

The bootloader program does not recognize that it's plugged into the preamp board simply by plugging in the usb cable. You have to put the preamp into bootloader mode. The preamp only enters bootloader mode when power is applied. When you apply power to the board it starts the bootloader code  within the board which then waits for only 5 seconds. During those 5 seconds the companion bootloader program running on your PC will see the preamp and you have to press the connect button on the app to lock in that connection. If you don't, the preamp terminates the bootloader code. If you miss the 5 second window you have to pull power from the board, wait a few seconds and then plug power back in.

I did get the bootloader program going. At first I failed to notice the "detailed instructions"and just plugged in the USB cable. 

 I had a bit of trouble getting the bootloader to start. It took at least three tries. It seemed like you had maybe 3 seconds to  press the connect button. Even if I pressed the connect button within the 5 second time period, the program didn't load consistently.  Nevertheless,  it did load and the download was successful.  Next time it'll be easier for me.

[tortugaranger]

There's a growing body of evidence that the improved accuracy of the autocal algorithm makes the quality of the power supply to the LDR3x board very important. During autocal the LDR3x board is acting as a closed loop measurement system involving 4 12 bit DAC channels, 3 op amps, and 4 analog-to-digital converter channels. So no big surprise that the better (cleaner, less noise) the PS, the easier it is for autocal to function properly.

Once upon a time I said that you could run the LDR3x off of most any 12V wall-wart style PS and while that remains true for normal running of the LDR3x board, it's not true when it comes to autocal where the presence of noise is more critical. With adjustable impedance relying heavily on autocal to work properly a clean power supply is more important than ever.

Going forward I am recommending that the LDR3x be powered either with a regulated linear power supply, or a high quality switch-mode power supply either of which are capable of supplying 12 VDC with 20 millivolt or less of ripple/noise and certainly none of the wild voltage spikes that are typical of many cheap SMPS "wall-wart" style power supplies. High frequency (above 100 kHz) low level noise from SMPS's is not a problem. We use a nice little SMPS dc-dc converter that produces clean +/- 10 millivolt power so it's not a linear vs. SMPS issue. It's a clean vs. noisy power supply issue.

[Tomy2Tone]

Just received my LDR3v2 and would like to understand the charts that robertopisa and Morten have supplied. My current source has about 100 ohm output impedance and one amp has 20k input impedance and another amp I have has 50k input impedance.

Anybody know what my impedance # on the LDR3v2 should be selected to? And how do you come to that?

Thanks,
Tom

[TJHUB]

I suggest setting the 5 presets to 10, 20, 40, 60, and 80 like I did.  You can run through them while listening to your most revealing and familiar tacks to hear the best setting.

[Tomy2Tone]

I suggest setting the 5 presets to 10, 20, 40, 60, and 80 like I did.  You can run through them while listening to your most revealing and familiar tacks to hear the best setting.

Ok, I've done some of those with 20 sounding the best so far..will the others. Thanks!

[justubes]

How does impedence selection work, for example the cdp xlr output impedence is 40 ohm and 4.8v rms.

I have measured a resistor at just before the xlr plugs at 10 ohm and 0.2ohm to ground, technically running this into the tortuga , what are the best resitors values?

Can and should the 10 ohm resistor be completely removed, or will increasing the value to 100 ohm or higher be more optimal into the tortuga?

This should be the i/v stage resistors at the cdp output?

I'm a little lost in what impedendece setting does.

[tortugaranger]

How does impedence selection work, for example the cdp xlr output impedence is 40 ohm and 4.8v rms.
I have measured a resistor at just before the xlr plugs at 10 ohm and 0.2ohm to ground, technically running this into the tortuga , what are the best resitors values?
Can and should the 10 ohm resistor be completely removed, or will increasing the value to 100 ohm or higher be more optimal into the tortuga?
This should be the i/v stage resistors at the cdp output?
I'm a little lost in what impedendece setting does.

While I'm not familiar with all the resistor options you apparently have with your CDP what matters is achieving a sufficient impedance bridging ratio between you CDP and you amp such that the CDP doesn't have to work all that hard (i.e. supply much current) in order to the deliver the audio signal (i.e. the AC voltage) to the amp. If your amp has say a 10k input impedance and your CDP has 10 ohm output, that's a 1000:1 ratio. Increasing the CDP to 100 out gives you 100:1.  I would argue anything beyond 50:1 is more than sufficient. Increasing it further is unlikely to yield any improvement in sound quality.

Putting a resistive passive preamp between your source and amp complicates matters because the preamp and amp act as parallel resistors such that their combined resistance is always less than each by itself. The adjustable impedance feature introduced with version 2.1.X firmware for the Tortuga preamps provides some degree of latitude to affect this overall combined impedance to optimize sound quality. If you already have sufficient impedance bridging I would argue that the adjustable impedance feature will not likely improve matters.

The recent article we published on Adjustable Impedance provides some fairly good background on this subject. Here is the article: http://www.tortugaaudio.com/ldr-passive-preamps-with-adjustable-impedance/

Cheers,
Morten

[tortugaranger]

A few thoughts on adjustable impedance.

My own impressions from using the adjustable impedance feature at various impedance levels (at least with our development audio gear) is that I can't discern much if any difference. The SS amp we use has a 20k input impedance and the LDR preamp has a nominal impedance input of 20k. Below 10k I can tell things start to lean out but going to 30, 40 or 50k seems to make no difference. I did convince myself that jumping it up to 99k did improve things just a bit.

I've heard back from a couple of customers with balanced units where pushing up the impedance much above 50k begins to exacerbate any DC offset that may be present and thus can introduce hum or even cause amps that have DC offset protection to trip. Keeping in mind that with balanced units 50k is really 100k and 99k would be 198k. This suggests that too high of an input impedance carries its own set of issues. DC offset may benefit from lower impedance where more DC signal current gets shunted to ground.

I suspect that once you achieve a sufficient bridging impedance ratio (amp input Z vs. source output Z), increasing it doesn't really buy you additional benefit.

I'm inclined in the next firmware release to leave in adjustable impedance but have only 1 setting and not 5. This would free up some code space for some other things that need updating. We're beginning to bump into the upper limits of the chip's programming capacity so have to be more judicious about what goes in and what might have to come out.

If any of  you have tried the adjustable impedance feature I would encourage you to share you thoughts.

Cheers,
Morten

[glynnw]

Like you, I heard little or no difference going into a tube amp. and maybe... just maybe... 99 sounded a bit tighter.

[robertopisa]

I agree with you. I am using 2K (which is 4K) with my balanced unit. This was better than 10k (which is 20k). Also 1 setting is enough for me. After I decided which is the right impedance, it is just a waste to keep 5 settings. The important feature is that we can change impedance in that setting.

The space you get free could be used to improve the ergonomics of the Apple remote. Too many functions overloading it. Maybe you should activate them only after pressing the encoder for 30 seconds. Normal use would leave just the basic functions (raise & lower volume, input selection, balance, mute). Just my 2 cents

A few thoughts on adjustable impedance.

My own impressions from using the adjustable impedance feature at various impedance levels (at least with our development audio gear) is that I can't discern much if any difference. The SS amp we use has a 20k input impedance and the LDR preamp has a nominal impedance input of 20k. Below 10k I can tell things start to lean out but going to 30, 40 or 50k seems to make no difference. I did convince myself that jumping it up to 99k did improve things just a bit.

I've heard back from a couple of customers with balanced units where pushing up the impedance much above 50k begins to exacerbate any DC offset that may be present and thus can introduce hum or even cause amps that have DC offset protection to trip. Keeping in mind that with balanced units 50k is really 100k and 99k would be 198k. This suggests that too high of an input impedance carries its own set of issues. DC offset may benefit from lower impedance where more DC signal current gets shunted to ground.

I suspect that once you achieve a sufficient bridging impedance ratio (amp input Z vs. source output Z), increasing it doesn't really buy you additional benefit.

I'm inclined in the next firmware release to leave in adjustable impedance but have only 1 setting and not 5. This would free up some code space for some other things that need updating. We're beginning to bump into the upper limits of the chip's programming capacity so have to be more judicious about what goes in and what might have to come out.

If any of  you have tried the adjustable impedance feature I would encourage you to share you thoughts.

Cheers,
Morten