Upgrading resistors?

[swich401]

Hi Bottlehead team,

What type of resistor are the 470 ohm 1/8W resistors attached to the tube pins?

What about the 100K ohm resistors on the active load boards, what type are they?


Any point replacing any of these resistors (or any other ones in the Mainline) with more "premium" resistor alternatives? Or are sound differences due to different types of resistors moreso "audiophoolery"?

I.e. for a more laid-back sound, would carbon film Kiwame 2 watt resistors work in the spots of the 470 ohm and 100K ohm resistors? Would they sound any different?

[Paul Birkeland]

Any point replacing any of these resistors (or any other ones in the Mainline) with more "premium" resistor alternatives? Or are sound differences due to different types of resistors moreso "audiophoolery"?

The 470 Ohm resistors are chosen mostly to fit.  If you try to put 2 Watt resistors in their place, reliability will be severely impacted.  The sonic impact of these resistors is also pretty minimal.

The 100 resistors on the active load board are completely out of the signal path.  Some people have replaced them, but I can't imagine why. 

Have Texas Components make you a pair of laser trimmed naked foil 14.7K resistors for the attenuators.  That resistor is always in the signal path.

-PB

[swich401]

For the 470 ohm resistors, will reliability be impacted mainly because of the size difference, or because of an electrical property of replacing those resistors with carbon film or tantalum alternatives? The 2W kiwame's are only 12mm body length, they should fit okay at an angle

Would replacing the 90.9 ohm dale resistors attached to the 6C45's cathode with 91 ohm carbon film or tantalum resistors have a sonic impact on the signal path? Any problems with doing this?

Also, is it okay to replace the 14.7K resistors of the attenuators with 15K resistors?

[Paul Birkeland]

For the 470 ohm resistors, will reliability be impacted mainly because of the size difference

Yes.  These are connections that you don't want to short to anything nearby.

The 2W kiwame's are only 12mm body length, they should fit okay at an angle

A 2W part is insanely overrated in this position.  There's no need to even use an 1/8W part.  Additionally, the sonic impact of these stoppers is quite minimal (unless they are removed).

Would replacing the 90.9 ohm dale resistors attached to the 6C45's cathode with 91 ohm carbon film or tantalum resistors have a sonic impact on the signal path? Any problems with doing this?

90.9 Ohms isn't an R1 value in the Mainline, but changing the R1 value on the B side will have a little impact that's noticeable.  This value is pretty tight, so do your best to get as close as possible (otherwise the sonic improvement you'd expect from the resistor change can be undone by an operating point shift).

Also, is it okay to replace the 14.7K resistors of the attenuators with 15K resistors?

I reran the numbers for you, and 15K is close enough that you shouldn't notice much.  The fine step sizes will move around a little bit, but not enough to be alarming. 

[swich401]

90.9 Ohms isn't an R1 value in the Mainline, but changing the R1 value on the B side will have a little impact that's noticeable.  This value is pretty tight, so do your best to get as close as possible (otherwise the sonic improvement you'd expect from the resistor change can be undone by an operating point shift).

Woops, I mean't the 90R9 Dale resistor hooked up to pin 7 of the 6C45's, which is the anode not the cathode.

What happens if I replace the 90R9 with a 91 ohm tantalum resistor? Any foreseeable problems?

I was thinking of the 2W for the 470 ohm resistors as that is the minimum that the Kiwame's come in. I may instead use 1/2 Watt tantalum resistors for the 470 ohm position, but as you said, I won't expect them to have much of an audible effect.

Thanks in advance for all the replies!

[Paul Birkeland]

Oh, yes, the plate stopper.  You can use a higher wattage part there since there's more space. 

[Doc B.]

Some have theorized that the smaller the resistor body the less susceptible it is to RFI pickup, so in theory it might be better to use a relatively small resistor for a stopper.

[swich401]

So there is no problem replacing the 90R9 resistor attached to pin 7 of the 6C45's with a 91 ohm resistor? What is the needed wattage for this resistor?

[Paul Birkeland]

So there is no problem replacing the 90R9 resistor attached to pin 7 of the 6C45's with a 91 ohm resistor?

Shouldn't be.

What is the needed wattage for this resistor?

0.17W

[swich401]

Thanks for all your help CB and Doc !!

I decided to go with 2 of each of the following TX2575 naked Z-foil resistors custom made by Texas Components:

90R9
14K7
49R9
16K2
5K76
2K05

Should target most of the sonically important signal path resistors correct?

[Doc B.]

Should target most of the sonically important signal path resistors correct?

At least in theory. But without trying both the stock resistors and the new resistors it would be impossible to say with certainty.

[Paul Birkeland]

16K2
5K76
2K05

Should target most of the sonically important signal path resistors correct?

No.  I wouldn't replace any of the attenuator resistors besides the 14.7K, and I certainly wouldn't replace half the resistors in the series string on one attenuator. 

[swich401]

No.  I wouldn't replace any of the attenuator resistors besides the 14.7K, and I certainly wouldn't replace half the resistors in the series string on one attenuator. 

Why so? Just due to cost (and too small of sonic influence of those coarse attenuator resistors) or because there would be some problem associated with replacing only half the resistors of the coarse attenuator for each channel?

My thought was that if I have the coarse attenuator set most often to -27dB then the 16K2, 5K76 and 2K05 resistors are in the signal path, whereas the 715, 255 and 140 ohm resistors wouldn't be as important sonically.

[Paul Birkeland]

My thought was that if I have the coarse attenuator set most often to -27dB then the 16K2, 5K76 and 2K05 resistors are in the signal path, whereas the 715, 255 and 140 ohm resistors wouldn't be as important sonically.

The signal path is a loop.

-PB

[NightFlight]

The closest thing in the Texas Components prefabricated line-up are the TX2352 (15K ohm @ 0.1%) rated to 0.4 W rms (0.6 W peak).  Without crunching numbers, I'm assuming this is more than good enough in the signal loop?