Hook-up Wires

[Mark]

While every one is adding their two cents, I'll add mine. The other factor people forget is memory. How long can you remember what the original amp sounded like compared to the modified amp?

I'll bet in 99.9% percent of cases people can't remember. When I do side by side evaluations I use my camcorder so I can quickly compare the two sounds. Also a device like a TC Ditto is handy as you're input a repeatable signal which is meanful to you.

However, I did have an instance where I heard a difference and it didn't come across on the camcorder. I suspect the camcorder was picking up the sound of the room over change in the amp.
This brings up the other question can your equipment measure the change. Another instant of this was a guy trying to measure the AC bias (30K+) voltage going to a recording head, his Fluke 77 told him there was no bias voltage. The trouble was the Fluke only read up to 8Khz (if I remember correctly).

Anyone want to talk about the differences between caps, resistor?

[Leo_Gnardo]

How long can you remember what the original amp sounded like compared to the modified amp?

Reading up on proper double blind testing techniques for audio - in the 1970's so this is nothing new - the ear's memory is 30 seconds. At best. Of course immediate switching between test items is ideal, but if there must be a pause make it brief as possible.

[martin manning]

Here’s some measured data:

Wire Capacitance Test using B&K 879B LCR meter
UL 1015 600V 22 AWG PVC insulation
Wire to ground plane:
18” wire laid on 12” sq. aluminum foil (curved so that the entire length is in close proximity to the foil surface), 24pF, 16 pF/ft or 1.33 pF/in.
Two samples, one with Brown and one with Purple insulation measured the same ;^)
Wire to wire:
Same two 18” wires loosely twisted (4 turns, so basically parallel), 16 pF,
10.66 pF/ft or 0.89 pF/in.
Same two wires twisted (11 turns) 17 pF, 11.3 pF/ft, 0.94 pF/in

For short wire runs, only a few pF of parasitic capacitance is involved. Longer runs can increase this to 10-20 pF, but in the worst cases (say a 1M volume wiper connected to a grid via 9” of 30 pF/ft shielded cable), the resulting corner frequency is still up around 27kHz. Note that Miller effect (mentioned above) applies only to capacitance between the grid and anode.

An A-B / B-A direct comparison seconds apart is the only listening test I give any credibility to. If minutes or hours pass in between, the aural memory is lost, IMO.

I’m looking forward to hearing about the results of an A-B comparison of holy grail plate resistors vs. current production Marcus is planning, using NOS resistors supplied by CW!

[shoggoth]

You don't need to dive into "did you remember the sound right" - there's +/- 20% variability in value on the caps in two otherwise identical builds, and probably different tubes. There are guaranteed audible, explainable differences, I don't know why anyone would decide wires were the cause.

Was going to volunteer to perform the experiment, signal generator -> 2 wires -> 2 inputs of differential amplifier -> coupling cap -> scope. If it's a straight line on the scope there's no difference. It would be a hand-built test jig though, and I don't know if any wire-audible-difference proponent would care about results, so unless a wire guy says "Do it and I will take the results seriously, or at least try to replicate and disprove the experiment", I'll hold off.

[martin manning]

Seems like this would be a good method to look for differences in components like caps and resistors too, but you might need to provide for a significant bias voltage to simulate the conditions in a tube amp. I would run the output of the diff amp into a spectrum analyzer; if your scope does that then great. If you see something the question of can you hear it is still unanswered. Unfortunately the only way there is a listening test.

[shoggoth]

My scope is a 1950's era Heathkit that I rebuilt. Straight lines is all I can do at the moment. Given that it would also be a hand-built differential amplifier (e.g. "best" op amp I can find in my parts box with 1% resistors and trimmer pots to zero stuff out during calibration), and that the eye is pretty good at seeing very small changes, it's probably sufficient.

If I wanted to do a spectrum analysis, I would have to use the PC.

I can test the jig with an RC circuit of known frequency roll-off instead of a teflon-coated wire and make sure the circuit responds to very tiny changes appropriately.

Anyhow, before I go through hours of labor I'd want to hear that a wire aficionado cared about the test, first.

[pompeiisneaks]

And he's right in saying that there is a lot of evidence sociologically that proves that people are really good at convincing themselves of things that can't actually be true, just due to their pre conceived bias. It doesn't mean you're not right, but unless you can measure it, it's nothing more than opinion instead of fact. No shame in that game, just come with data my friends

~Phil

Maybe so Phil but be careful with this because it can go either way. The "science" guys are so convinced that their graphs and simulations tell the whole story that they won't even bother to listen for themselves. So convinced that it can't be true that they would deny it even if they heard it. I guess everyone has their own way of doing things but my hearing is the most important tool I use when I build or repair amplifiers. I don't really have time nor feel the need to build and rebuild an amplifier to prove something I have already experienced to someone else. What pissed me off was that the only guy on this thread that even mentioned listening to the wire was verbally bullied off the thread.
CW

I am not discounting hearing being a great measure, and as has been mentioned, go with what sounds right, but for it to be reproducible, you need to understand the why, not just say unequivocally that it is.

I'm going to make up something, completely artificial, as a 'what if' situation.

What if someone built amps with a specific kind of wire and were convinced that when they used it, it improved sound, so they always used it thereafter and were confident, in fact they could PROVE by old amps vs new amps 'sound' that they were improved. BUT let's also say it was their 20th amp, and at that point they basically unintentionally had caught one of the last bad habits they have of making connections 'less than stellar' so really every amp they made after that point had better quality connections, but they didn't put 1 and 1 together.

In a nutshell, with so many massive variables in amp construction, 'knowing' something to be fact is very hard, and requires pretty dedicated scientific approach, imo. And I'm still a major neophyte when it comes to amps, but I've been doing engineering work all of my life, early on it was in construction/civil engineering, mid way through it was computers til now. I deal with massive amounts of variables on linux servers/systems. If someone there was to say the interconnects of the CAT6 cable were what caused the problems in the network, and couldn't prove it was just a bad batch somehow, they'd be laughed out of the job.

Another one that drives me nuts are how ridiculously expensive they make HDMI cables.

It's a DIGITAL signal. Either the 0's and 1's make it across, or they don't if they don't you get 100% nothing, useless information. It's either on, or off, but they still convince people to buy gold plated, directional, touched by virgins, cryogenic-ally treated HDMI cables that work identically to a 7$ one.

In a nutshell, I'll never mock someone for their opinion, I just won't consider it as fact until proven, but I even act on some peoples opinions based on the respect I have for them. They may know something I don't and I'll find out later why it works, if I can.

Hope that makes sense?

I did mention that I felt that R.G. could have done it nicer, and he even apologized for going too far, but nowhere did I see any name calling or outright insults, so for me, that's passable enough for a civil discussion.

~Phil

[Bombacaototal]

Resurrecting this topic with a couple of questions.

1) I am planning on doing a build with a single colour hook up wire (as it will be cheaper to get a big reel in one colour). Is it a bad idea in terms of potential for error or for tracing cables back? I think for heaters I will use a different set for the colour code just to be on the safe side...

2) Is 3A current rating enough for hook-up wires a Dumble SSS 50W style amp? I think 3A is usually suitable for use throughout the whole circuit for most 5-40W amp designs, but what if I use 3A on all hook-up wires except the heaters?

[pompeiisneaks]

Resurrecting this topic with a couple of questions.

1) I am planning on doing a build with a single colour hook up wire (as it will be cheaper to get a big reel in one colour). Is it a bad idea in terms of potential for error or for tracing cables back? I think for heaters I will use a different set for the colour code just to be on the safe side...

2) Is 3A current rating enough for hook-up wires a Dumble SSS 50W style amp? I think 3A is usually suitable for use throughout the whole circuit for most 5-40W amp designs, but what if I use 3A on all hook-up wires except the heaters?

I don't think it's a big deal, fender often did yellow across the entire tube hookup section excepting the green for heaters. Yes it can make troubleshooting a bit trickier, but continuity testers are there for a reason

~Phil

[Mark]

The insulation has a voltage rating, have you taken that into consideration?

I would be more concerned about that, as most the triodes pass 1mA of current. The filaments and the output valves are the exception of course.

[Bombacaototal]

Thank you very much for the reply Phil, i am sure I will regret going one colour but I will give it a try

Thanks Mark, the ones I am looking at are 600V, 3A rated.