[ALRAINBOW]

Yes thanks . It's just barely warm now. I think Bruno and them were right it's not hot till your fingered sticks to it. LOL.

 

[m17xr2b]

Just finished my test with the resistor box using the TBI Millenia and LCD3. I bought about 50 resistors to test all possible configurations. With the optimal resistor box(R2 = 6, R3 = 2.2) the hiss is completely gone but the dynamics of the music are missing, the high notes are a bit silent, soundstage a bit smaller, basically everything is toned down. Using just attenuation so R1 = 147(to match the attenuation of the resistor box) is sound much more better. From initial listening tests it sound 95% like using the LCD3 directly from the taps but without the hiss. Currently I am using R1 for 100ohms and I think will be my setup of a while. 
I am curious why adding two resistor changes the sound so much as opposed to just one.  
My concern is if the Millenia can handle this long therm since it's effectively sees 150ohm speakers. Also I had some initial worries about the damping factor but it looks like it does not affect planar magnetic drivers as much as I thought.

 

[john57]

  Just finished my test with the resistor box using the TBI Millenia and LCD3. I bought about 50 resistors to test all possible configurations. With the optimal resistor box(R2 = 6, R3 = 2.2) the hiss is completely gone but the dynamics of the music are missing, the high notes are a bit silent, soundstage a bit smaller, basically everything is toned down. Using just attenuation so R1 = 147(to match the attenuation of the resistor box) is sound much more better. From initial listening tests it sound 95% like using the LCD3 directly from the taps but without the hiss. Currently I am using R1 for 100ohms and I think will be my setup of a while. 
I am curious why adding two resistor changes the sound so much as opposed to just one.  
My concern is if the Millenia can handle this long therm since it's effectively sees 150ohm speakers. Also I had some initial worries about the damping factor but it looks like it does not affect planar magnetic drivers as much as I thought.

Last night I saw some charts made with different impedance loads for some class D amps. I noticed that lowering the output impedance loading with the two resistor model that the amp sees have a tendency to have more of a high frequency roll-off  at 10kHz than with higher impedance loads. This going inline to what your saying for your particular amp. I think that damping factor with headphones have less impact as compared to the much heaver drivers of speakers.

 

[potterma]

  Just finished my test with the resistor box using the TBI Millenia and LCD3. I bought about 50 resistors to test all possible configurations. With the optimal resistor box(R2 = 6, R3 = 2.2) the hiss is completely gone but the dynamics of the music are missing, the high notes are a bit silent, soundstage a bit smaller, basically everything is toned down. Using just attenuation so R1 = 147(to match the attenuation of the resistor box) is sound much more better. From initial listening tests it sound 95% like using the LCD3 directly from the taps but without the hiss. Currently I am using R1 for 100ohms and I think will be my setup of a while. 
I am curious why adding two resistor changes the sound so much as opposed to just one.  
My concern is if the Millenia can handle this long therm since it's effectively sees 150ohm speakers. Also I had some initial worries about the damping factor but it looks like it does not affect planar magnetic drivers as much as I thought.

Curious observation.  Were all of the resistors the same, besides resistance value?  Are you sure you had good, tight connections with the multi-resistor configuration? 
 
The Millenia won't care about the load its driving, from a damage-the-amp standpoint. You are correct about the planar's not being as sensitive to damping factor as dynamic 'phones.
With the 6 and 2.2 ohm configuration, you are demanding much more current from the amp for a given SPL level, though, so maybe that is part of the change you heard?
 
@john57 Do you have a link to those plots?  I think we all would be interested to see them.  Thanks in advance!

 

[zilch0md]

Someone can correct me if I'm wrong, but the Millenia MG3 would only be at risk if it saw less than a 4-Ohm load on 12V power, or less than 8-Ohms on 24V power - IIRC from reading the manual.
 
In general, amps can handle higher impedances, without difficulty.
 
As long as you have  a good solder connection, I think stacking two resistors serially would be OK, too.  
 
I found that I could still hear a little bit of hiss when using the LCD-2 on the MG3 with two 50-Ohm wirewound resistors stacked serially.  100-Ohms wasn't enough attenuation (-12 dB) to completely kill the hiss for the my 50-Ohm LCD-2.
 
And yes, when the LCD-2 is connected with -15.8 dB attenuation, the hiss is gone, but the dynamics are indeed compromised.  Still, the MG3, with only 133mW into 50-Ohms, after attenuation, has a wounderful, sweet, organic, sound with the LCD-2, that I don't have with the (more dynamic) 680mW into 50-Ohm CEntrance DACmini CX, for example.
 
Mike

 

[brunk]

I too have found similar findings with resistors, but my experience is from the HE-adapter box only at the moment 

 This makes me think that the resistors may play some sort of effect on the more demanding dynamic swings. Perhaps the resistors absorb these short peaks and are not efficient enough to pass it on? Perhaps this is where a carbon-based or some other type of resistor can shine vs. the wirewounds?
I love a good mystery 

 

[brunk]

@potterma perhaps you are onto something here in regards to the mechanical connection of the resistors. The losses we are experiencing may in fact be due to the leads of the resistors themselves. I may have a possible solution for that. Can you or someone else using the resistor network snip off the leads close as possible to resistor and solder on a 20awg solid core fly lead and see if listening impressions improve? Most resistor leads are quite thin and we may be running into a limiting/loss effect due to that. I would do it myself, but I only have the HE-adapter box and I really don't feel like hacking it up right now lol.

 

 

[potterma]

  @potterma perhaps you are onto something here in regards to the mechanical connection of the resistors. The losses we are experiencing may in fact be due to the leads of the resistors themselves. I may have a possible solution for that. Can you or someone else using the resistor network snip off the leads close as possible to resistor and solder on a 20awg solid core fly lead and see if listening impressions improve? Most resistor leads are quite thin and we may be running into a limiting/loss effect due to that. I would do it myself, but I only have the HE-adapter box and I really don't feel like hacking it up right now lol.

 

Awwww come on.... Where's your spirit of adventure?
 
I would think that the resistor-resistor connection and the resistor type would have more effect than the resistor leads.  I tried metal film power resistors and did not like them at all.  Killed the hf dynamics, for some reason.
 
I might have to make some carbon comp resistors and give them a try.  Can't stomach paying for 'em, though :choke: :cough:

 

[brunk]

  Awwww come on.... Where's your spirit of adventure?
 
I would think that the resistor-resistor connection and the resistor type would have more effect than the resistor leads.  I tried metal film power resistors and did not like them at all.  Killed the hf dynamics, for some reason.
 
I might have to make some carbon comp resistors and give them a try.  Can't stomach paying for 'em, though :choke: :cough:

Lol, yeah I agree on both adventure and effect (lack of) of resistor leads, but its a cheap non-destructive option. Hehe! Yeah those things are quite expensive for what is effectively a pencil lead 

 

[potterma]

  Lol, yeah I agree on both adventure and effect (lack of) of resistor leads, but its a cheap non-destructive option. Hehe! Yeah those things are quite expensive for what is effectively a pencil lead 

Not just "effectively a pencil lead".  EXACTLY a pencil lead...

 

[brunk]

  Not just "effectively a pencil lead".  EXACTLY a pencil lead...

You got me there! 

 

[Chris J]

i have figured out my resister heat problem . I just put two 20 ohms in parralell so all is better now. with the amp i am using now i need the attenuation and proper loading of amp output. with the he6 its fine with no attenuation but there is a slight low hiss and with the HD800 the his is there but attenuation is a must to get to 1/3 of the volume level and not be very loud. thanks so much .

al d

Excellent news Al!
Sounds like you have solved the problem by splitting the power dissipation between two resistors?

 

[ALRAINBOW]

Yes it is , but I really wanted to know why I was approaching the ten watt rating in the first place . Lol.

 

[Chris J]

Just finished my test with the resistor box using the TBI Millenia and LCD3. I bought about 50 resistors to test all possible configurations. With the optimal resistor box(R2 = 6, R3 = 2.2) the hiss is completely gone but the dynamics of the music are missing, the high notes are a bit silent, soundstage a bit smaller, basically everything is toned down. Using just attenuation so R1 = 147(to match the attenuation of the resistor box) is sound much more better. From initial listening tests it sound 95% like using the LCD3 directly from the taps but without the hiss. Currently I am using R1 for 100ohms and I think will be my setup of a while. 
I am curious why adding two resistor changes the sound so much as opposed to just one.  
My concern is if the Millenia can handle this long therm since it's effectively sees 150ohm speakers. Also I had some initial worries about the damping factor but it looks like it does not affect planar magnetic drivers as much as I thought.

It does not matter to the amp if it seems a continuous 150 Ohm load.
SS amps do not have stability problems with high impedance loads.
You are more likely to run into trouble when an amp is driving a load with too much capacitance.

Several years ago I noticed my Audiolab 8000P power amp was running very hot, turns out the amp was oscillating because the loudspeaker cable I was using was too capacitive.
An oscilloscope hooked up to the speaker terminals showed a constant 200 or 300 kHz sine wave at the amp's output.

 

[ALRAINBOW]

Now that is interesting . I have read of this happening but the truth is I never thought is was true. Please tell me what kind of wire and it must have been a long run , how far.?