DIY Line Out Cable - Works!

@skinjob wrote:


axon01 wrote: 

I’m curious about what you said about the line out. I’m new here and haven’t found much information about the nature of the line out on the Fuze.  For instance, why does the line out need 3 volts?


 

The 3 volts is not on the line out itself, but is necessary to enable the line out.  Ordinarily, the dock would supply the 3 volts on a certain pin.  This tells the Fuzz it’s connected to a dock and should enable it’s liniment and disable it’s headphone out.

 

As to the other comments about SQ through the dock, it will be interesting to see how much difference there is between a simple cable with no extra electronics versus the dock where the signal passes through another circuit board.

Hi Skin and Neutron

It will also be interesting to compare the signals from a line out cable, a high quality dock and the headphone jack.

From what I am beginning to understand, the Fuzz is applying it’s amplification (or attenuation), digitally, through the chip, (thank you Neutron).  It makes sense then, that when the dock is deployed, the headphone jack is not, because the signal from the chip is routed to either the jack or the line out. 

The fact that you are able to get a signal from either the line out or headphone jack but not both and the position SanDisk placed the headphone jack, (intentionally precluding use while docking), seems to bear out that there are not two amplifiers, as I have read. 

The digital concept makes further sense because there is no distortion regardless of the volume level.  The interesting thing is that if Neutron’s supposition is correct,  that the gain is being applied digitally by the chip, the exact same signal would be available through the jack or the line out.   This, btw, agrees with what I heard when comparing the Griffin dock to the headphone jack.  The headphone jack, having less to go through, had the cleaner signal. 

More digital evidence might be when someone reported that when they used the remote on the sansa dock to change the volume, it did so by changing the volume on the Fuze.

axon

@14124all wrote:

I have to agree with Skinjob and neutron_bob. The Fuze is implementing the line out in it’s own way. It should not be a concern about if it is a “true” line out or not. This was probably based on home stereo equipment with a 1volt peak signal. Are car stereos true line out with the 5volt pre-out signals?

The main thing is that there is a high quality, high impedance signal available to drive pre-in inputs harder than the headphone out. If I run the Fuze volume all the way to max, the signal is noticeably lower than with the Griffin Dock or the cable mod. The purpose of this project was to get a signal that is more useable with car stereos and line-in home equipment, as the headphone out does not drive them well. It seems to me that it works quite well without signal degradation, so I am happy with it.

If you are looking for audiophile quality, true to spec line out, you probably should not be using a compressed audio player anyway!

Hi 14

This is really strange.  I don’t know why this was not what I found?  Fuze volume set to 100% gave me the same volume from the Griffin dock and the headphone jack.  I used an 18" Zu  Pivot (mini to RCA) to my M^3 and DT880s for the comparison.  

 

axon

 

Edit to add link:  What I used 

Message Edited by axon01 on 12-02-2008 11:29 AM

@axon01 wrote:


@14124all wrote:

I have to agree with Skinjob and neutron_bob. The Fuze is implementing the line out in it’s own way. It should not be a concern about if it is a “true” line out or not. This was probably based on home stereo equipment with a 1volt peak signal. Are car stereos true line out with the 5volt pre-out signals?

The main thing is that there is a high quality, high impedance signal available to drive pre-in inputs harder than the headphone out. If I run the Fuze volume all the way to max, the signal is noticeably lower than with the Griffin Dock or the cable mod. The purpose of this project was to get a signal that is more useable with car stereos and line-in home equipment, as the headphone out does not drive them well. It seems to me that it works quite well without signal degradation, so I am happy with it.

If you are looking for audiophile quality, true to spec line out, you probably should not be using a compressed audio player anyway!


Hi 14

 

This is really strange.  I don’t know why this was not what I found?  Fuze volume set to 100% gave me the same volume from the Griffin dock and the headphone jack.  I used an 18" Zu  Pivot (mini to RCA) to my M^3 and DT880s for the comparison.  

 

axon

 

Edit to add link:  What I used 

Message Edited by axon01 on 12-02-2008 11:29 AM

Hmm, not sure why you’re not hearing a noticable difference.  It’s fully expected that a headphone-out connected to a normal line-in will result in a low level.  The difference is usually very obvious.  Is it possible the line device you are plugging into has some kind of auto-sensing jack and is compensating for the headphone input?  A lot of newer devices that were designed with portable gear in mind handle headphone input better than devices designed for traditional home a/v gear.

axon01, take a look at the block diagram for the AMS AS3525 SOC. It may answer some of your questions. The system chip has 3 independent audio outputs on chip: headphone amp, line out and speaker amp. All of their inputs are driven by a common signal with the DAC and AGC limiter upstream. Easily manipulated digitally before the DAC output. Note that the line out output does not feed the headphone out input. This may clear up some confusion.

Google for AustriaMicrosystems AS3525 to find the function sheet and block diagram. Sorry, I do not have a URL handy.

WhiiteRaven, the AMS chip does have 2 dedicated line inputs. There most likely is no hardware connection on the Fuze board to the chip pins and no enable in the firmware. Maybe the next generation Fuze or another model will enable it. This chip is loaded with features that are not used in the Fuze.

@skinjob wrote:


@axon01 wrote:


@14124all wrote:

I have to agree with Skinjob and neutron_bob. The Fuze is implementing the line out in it’s own way. It should not be a concern about if it is a “true” line out or not. This was probably based on home stereo equipment with a 1volt peak signal. Are car stereos true line out with the 5volt pre-out signals?

The main thing is that there is a high quality, high impedance signal available to drive pre-in inputs harder than the headphone out. If I run the Fuze volume all the way to max, the signal is noticeably lower than with the Griffin Dock or the cable mod. The purpose of this project was to get a signal that is more useable with car stereos and line-in home equipment, as the headphone out does not drive them well. It seems to me that it works quite well without signal degradation, so I am happy with it.

If you are looking for audiophile quality, true to spec line out, you probably should not be using a compressed audio player anyway!


Hi 14

 

This is really strange.  I don’t know why this was not what I found?  Fuze volume set to 100% gave me the same volume from the Griffin dock and the headphone jack.  I used an 18" Zu  Pivot (mini to RCA) to my M^3 and DT880s for the comparison.  

 

axon

 

Edit to add link:  What I used 

Message Edited by axon01 on 12-02-2008 11:29 AM


 

Hmm, not sure why you’re not hearing a noticeable difference.  It’s fully expected that a headphone-out connected to a normal line-in will result in a low level.  The difference is usually very obvious.  Is it possible the line device you are plugging into has some kind of auto-sensing jack and is compensating for the headphone input?  A lot of newer devices that were designed with portable gear in mind handle headphone input better than devices designed for traditional home a/v gear.

Hey Skin and 1412

I ran some comparisons today in my office using the Griffin dock connected to the tape loop out on a mini component unit I bought in the 80’s.  There was no obvious difference between volumes of the headphone out jack and the dock jack, with the Fuze set to 100%. 

For some reason, I am not getting the expected results.  I had more than one person listen as I switched back and forth and no one seemed to feel there was an obvious volume difference. 

This bears out what I was hearing from my headphone amp at home. 

This is not the case, however, if the Fuze is set to less than 100% volume.  In that scenario, the dock is noticeably louder than the Fuze, by degrees, as the Fuze approaches 100% volume. 

I’ll try it again tomorrow, but I’m very confident in what I’ve heard, through both my headphone amplifier and now my office stereo.

I also discovered that (while listening to speakers), I was able to notice a slight increase in volume as I dialed up from 50% to 100% with the Fuze plugged into the dock.  This is not an obvious volume change, but it is detectable.  I only mention this because previously I thought there was no volume increase from 50% upward.

axon

axon, See what version firmware is on your player. If it is the 07 version try upgrading to the 11 version and retry your testing at 100% volume. I need to further test, but I have players with both versions and they do react differently. The volume control acts different when on my lineout cable between these 2 versions. I will verify with the Griffin Dock what happens (it should still be the same as my cable) and get back to this thread. I don’t use the current firmware version 15 due to playback problems encountered with it. Version 11 seems to work the best for audio playback, which is all i’m concerned with.

axon, WOW!!! Disregard previous post. There is a difference in firmware versions, but not with affected volume. What I found - You are correct!!! At 100% volume there is barely any differnce in volume output from the Griffin dock out or player headphone out. With my lineout cable there is a noticeable increase in volume! This renders the Griffin Dock rather useless as is for a lineout connection. This should be easy to alter the Griffin Dock to get the signal straight through. The Griffin Dock has series 220uf decoupling caps and parallel 10k impedance output resistors feeding the output jack. Time to bypass the resistors and maybe the caps!

NOTE: Another problem with the Griffin Dock is loss of the FM antenna. The player seems to only connect the FM antenna through the headphone jack common audio connection with no connection on the 30 pin connector. The Griffin dock has no connection for the antenna! Activating it will probably require an internal jump in the player to an unused pin on the 30 pin connector and a single wire to act as antenna on my cable. Getting the antenna to work might take some doing for other alternatives.

More finds! Sorry had a typo in the previous post. The output resistors are 100k not 10k. I disconnected them and had no difference in volume level, so it doesn’t seem to be impedance mismatch related. I’m going to reverse-engineer the Griffin board completely and come up with a full circuit diagram and hopefully see what is going on here. Since my cable does get a noticeable volume increase, it may be the voltage on pin7 may act as a gain control. Only speculation right now, I need to do more tinkering with the dock.

@14124all wrote:

axon, WOW!!! Disregard previous post. There is a difference in firmware versions, but not with affected volume. What I found - You are correct!!! At 100% volume there is barely any differnce in volume output from the Griffin dock out or player headphone out. With my lineout cable there is a noticeable increase in volume! This renders the Griffin Dock rather useless as is for a lineout connection. This should be easy to alter the Griffin Dock to get the signal straight through. The Griffin Dock has series 220uf decoupling caps and parallel 10k impedance output resistors feeding the output jack. Time to bypass the resistors and maybe the caps!

NOTE: Another problem with the Griffin Dock is loss of the FM antenna. The player seems to only connect the FM antenna through the headphone jack common audio connection with no connection on the 30 pin connector. The Griffin dock has no connection for the antenna! Activating it will probably require an internal jump in the player to an unused pin on the 30 pin connector and a single wire to act as antenna on my cable. Getting the antenna to work might take some doing for other alternatives.

Interesting.  I was going to get one of the Griffin docs while they were still on clearance, but if it doesn’t solve the level issue then there’s really no point.  Man, it really shouldn’t be this hard to get a decent line out.

I still haven’t had a chance to mod my cable, but I’m hoping to try this week.

Next release of Fuze firmware will output fixed line level when connect to the Dock.

I like aspects of either configuration, as one could tweak the volume while docked.

The advantage of fixed becomes apparent when I use the e200v2 on the dock: the volume setting for the headphone is retained, and restored when I pull the device from the dock.  Quite convenient!

Bob  :smileyvery-happy:

Doing some quick checks it appears that the voltage on pin7 of the Fuze controls the lineout level. The Griffin Dock feeds 5volt through a series 47k resistor and has 1 volt on pin7. My modded cable feeds 3 volts to pin7 and produces a louder output signal. The Griffin Dock can be modded for a higher than 1 volt feed with a reistor change, or a pot and resistor could be used to produce a voltage limited gain control. I would imagine that the feed should not exceed 3 volts, so values must be chosen correctly if modding the dock due to the 5 volt feed. I do not have a Sansa Dock, but it seems to be more involved internally than the Griffin Dock. I’ll be trying mods on the Griffin Dock in the next few days.

@skinjob wrote:


@14124all wrote:

axon, WOW!!! Disregard previous post. There is a difference in firmware versions, but not with affected volume. What I found - You are correct!!! At 100% volume there is barely any difference in volume output from the Griffin dock out or player headphone out. With my lineout cable there is a noticeable increase in volume! This renders the Griffin Dock rather useless as is for a lineout connection. This should be easy to alter the Griffin Dock to get the signal straight through. The Griffin Dock has series 220uf decoupling caps and parallel 10k impedance output resistors feeding the output jack. Time to bypass the resistors and maybe the caps!

NOTE: Another problem with the Griffin Dock is loss of the FM antenna. The player seems to only connect the FM antenna through the headphone jack common audio connection with no connection on the 30 pin connector. The Griffin dock has no connection for the antenna! Activating it will probably require an internal jump in the player to an unused pin on the 30 pin connector and a single wire to act as antenna on my cable. Getting the antenna to work might take some doing for other alternatives.


 

Interesting.  I was going to get one of the Griffin docs while they were still on clearance, but if it doesn’t solve the level issue then there’s really no point.  Man, it really shouldn’t be this hard to get a decent line out.

 

I still haven’t had a chance to mod my cable, but I’m hoping to try this week.

That’s no reason not to get a Griffin Dock.  I ordered a 2nd one for home use.  The convenience factor is too great not to have one and they’re practically free.   The SQ is from the dock is great, and I’m not sure if a firmware lineout can sound better.

axon

Message Edited by axon01 on 12-08-2008 08:51 PM

@sansafix wrote:
Next release of Fuze firmware will output fixed line level when connect to the Dock.

Hi sansafix

If the firmware is just going to eliminate the ability to control the volume of the dock from the Fuze, my suggestion is to leave it as is.  It’s a nice feature to have and since the signal is being controlled digitally, the end result sound quality will remain the same as it is now, whether we can change the volume or not. 

If the headphone lead from the DAC chip goes to the headphone jack and the lineout lead from the DAC goes to the docking plug, there is no reason to remove the ability to adjust the volume on the dock.  However, from my experience with the Fuze, I don’t believe that is the case.  ‘To simplify the construction,’ and the fact that firmware can alter this feature, leads me to think there is just one lead from the DAC that goes to the volume control and then either to the headphone jack or the docking plug, allowing the Fuze to control the volume of the headphone and the dock. 

This is not a bad thing, however, and my advice to SanDisk is to leave it alone.  Firmware is not going to change the source of the sound or the sound quality.  The Dockable Fuze is a winner just the way it is!  There is no reason to remove the volume feature just to make the Fuze have a fixed line level when connected to a dock.   That would be a downgrade rather than an upgrade.

axon

Message Edited by axon01 on 12-08-2008 08:53 PM

@neutron_bob wrote:

I like aspects of either configuration, as one could tweak the volume while docked.

 

The advantage of fixed becomes apparent when I use the e200v2 on the dock: the volume setting for the headphone is retained, and restored when I pull the device from the dock.  Quite convenient!

 

Bob  :smileyvery-happy:

Hi N_B

I don’t see the advantage of a fixed lineout.   Right now I can leave my headphone setting right where it is, dock the Fuze and get a line level signal for my office stereo, but I also can lower the line level out to the dock should I want to plug a sensitive headphone into it for listening at home.   And should I need to balance the line out level with my other components (which I don’t, it’s as loud as my tuner), I have the ability to, as you said above, tweak the volume while docked.

axon

axon, Now I’m confused! In 2 of your posts you say the headphone out seems cleaner than the dock out and the volume level is near equal. I thought your dock lineout level was low, maybe I misunderstood. So why bother with the dock? Just use a cable on the headphone out since the level is the same and the quality is better or equal and you retain full volume control. The dock also kills FM reception so why bother with it?

The general problem with applying the volume completely digital is the signal to noise ratio of the DAC. For example on a 16 bit DAC with a 16 bit signal the volume setting 25% would require to drop the signals lowest two bits making it in fact a 14 bit signal and so increasing its quantization noise (relative to the full signal).

If I read correctly the Fuze’s DAC has 18 bit, while all audio played has 16 bit. So if you would like to amplify as much as possible in software (digital) without introducing additional noise, then you would do the lower 25% of the amplification by leaving the digital signal untouched and just amplifying it with an analogue circuit. Beyond this point you can do the amplification in software by multiplying the signal with a factor of up to 4 without overflow and without introducing errors bigger than 0.5 digit.

This also relieves the hardware developer mostly of the burden of creating an analogue amplification circuit that is neutral in sound all the way from lowest to highest amplification. You just have to assure that the highest amplification is perfect, then 75% of the volume range is always perfect.

So what seems to happen in the current firmware when line out is activated is that the digital amplification is set to max, while analogue amplification still depends on the volume setting as before.

calv, Thanks for the excellent explanation! I have only dealt with DAC operation on industrial motor control and was not aware of the audio amp workings. What you have presented explains why the Griffin Dock output will increase from the Fuze volume control to about 25 or 30 percent and then sits at that peak with no more increase all the way up. My lineout cable is probably overdriving the pin7 input and that is why I get a noticeable volume increase over the dock. I’m not hearing audio distortion on a Klipsch Promedia (did hear some clicks at times which may be clipping), but it probably is there and would be heard on axon’s DT880s. I’ll drop the voltage feed to pin7 with a resistor instead of a straight jumper.

@14124all wrote:
calv, Thanks for the excellent explanation! I have only dealt with DAC operation on industrial motor control and was not aware of the audio amp workings.

Well, to be honest I have no practical experience with DAC or analogue circuits at all. Its just composed of what I think to have learned at university. Now I’m more of a computer scientist :slight_smile: so someone who actually knows the matter from real experience might correct me in one or the other thing.