200 hours

[Atmasphere]

I have often wondered, is 200 hours of break in for the component or for your brain/ear to come round.

There really is something to it. We can measure differences in power supplies after the break-in. We use electrolytic capacitors (no choice on that BTW); I was told years ago by a Cornell-Dublier engineer that the electrical charge on the electrolytic slowly forces water vapor out of the capacitor as it 'forms up' and becomes more efficient. One example of that that is easy to measure is the difference in current inrush. Early on, we've seen it blow fuses that later on there is just no way that fuse is going to blow. At the same time we can see higher power supply voltages. So filter capacitors forming up to the voltage at which they will be used is part of the process.

But that does not seem to be the whole picture. There seems to be something with wiring too- when we disturb an amp or preamp's wiring during an update, it seems that that unit is unhappy about it for a while. Sometimes it might take a few weeks to come around. Unfortunately, we've yet to sort out a way to measure what this phenomena is, although the fact that it is occurring is easy enough to measure if you simply put a microphone in the room and take some frequency response and distortion measurements, then compare 'before' and 'after'. But I would like a way that gets me a little closer to the cause.

With new equipment that first 24 hours is the period that we don't really want anyone but us to hear what the amp or preamp is doing. The sound is stilted and not really that impressive. As time goes by it does seem that overall the equipment gets more lucid. It does not seem to be simply the ear adjusting to the sound over time, although I am sure that occurs too. Again, I would really like to get a better handle on what break-in is all about. We've been at this for 38 years and deal with the phenomena on a consistent basis. Its irritating to be able to consistently hear something but not to have good tools to measure what that something is!

 

[Whatmore]

But that does not seem to be the whole picture. There seems to be something with wiring too- when we disturb an amp or preamp's wiring during an update, it seems that that unit is unhappy about it for a while. Sometimes it might take a few weeks to come around. Unfortunately, we've yet to sort out a way to measure what this phenomena is, although the fact that it is occurring is easy enough to measure if you simply put a microphone in the room and take some frequency response and distortion measurements, then compare 'before' and 'after'. iBut I would like a way that gets me a little closer to the cause.

Now this is interesting. I've never seen anyone post measurements of in-room differences caused by power cables. Do you have any?

 

[FrantzM]

Now this is interesting. I've never seen anyone post measurements of in-room differences caused by power cables. Do you have any?

Would like to see those too...

 

[Speedskater]

Now this is interesting. I've never seen anyone post measurements of in-room differences caused by power cables. Do you have any?

While in-room measurements with microphones would add a lot of uncontrolled variables, what would be best is measurements at the speaker terminals.

 

[Atmasphere]

Heck- power cables are easy. You don't even need in room measurements for that.

I've covered this on this site before, but I can do it again. We've seen a power cord and nothing else be responsible for a loss of power amounting to about 40 watts in one of our amps. All I did there was measure the voltage drop across the power cord and I found the culprit. I did that with a 3 1/2 digit DVM. That kind of loss being pretty measurable I would expect it to be audible too, and it is.

Can we just put up a sticky about why power cords are audible and measurable so we don't have to keep talking about it? I can supply the theory behind it on the electronics side; that is, why it make a difference to an amp or preamp.

 

[Speedskater]

..................................
I've covered this on this site before, but I can do it again. We've seen a power cord and nothing else be responsible for a loss of power amounting to about 40 watts in one of our amps. All I did there was measure the voltage drop across the power cord and I found the culprit. I did that with a 3 1/2 digit DVM. That kind of loss being pretty measurable I would expect it to be audible too, and it is.
.......................................

Yes, bad power cords are not uncommon!
One day I measured the IEC power cords in my spares box.
Using a 6½ digit 4 terminal Ohm meter. Some were way more than the calculations of length and AWG would lead you to expect.

 

[Speedskater]

............................... what would be best is measurements at the speaker terminals.

Thinking about it more. Most truly audible differences will be situation specific. Poorly designed audio component power supplies or nearby appliances or lighting systems will account for almost all the audible differences. So these many uncontrolled differences could mess-up the test results.

 

[Whatmore]

Heck- power cables are easy. You don't even need in room measurements for that.

I've covered this on this site before, but I can do it again. We've seen a power cord and nothing else be responsible for a loss of power amounting to about 40 watts in one of our amps. All I did there was measure the voltage drop across the power cord and I found the culprit. I did that with a 3 1/2 digit DVM. That kind of loss being pretty measurable I would expect it to be audible too, and it is.

Can we just put up a sticky about why power cords are audible and measurable so we don't have to keep talking about it? I can supply the theory behind it on the electronics side; that is, why it make a difference to an amp or preamp.

What voltage drop does a 40w change equate to?
Surely such a power cord is faulty and should be binned?

 

[Atmasphere]

A lot of power cords have similar drops. This is your typical 15 amp 18Ga cord. The voltage drop across the cord was between 2 and 3 volts- the DVM did not have the resolution to tell me more. I could set up the measurement again though. At the time, I just wanted to know why the amp was not making full power. Once I compensated for the loss on the variac, the power was back.

 

[Whatmore]

A lot of power cords have similar drops. This is your typical 15 amp 18Ga cord. The voltage drop across the cord was between 2 and 3 volts- the DVM did not have the resolution to tell me more. I could set up the measurement again though. At the time, I just wanted to know why the amp was not making full power. Once I compensated for the loss on the variac, the power was back.

This is quickly running into unkown territory for me, but a 3V drop with 15amps of current equates to resistance of 0.2 Ohms (did I do that correctly?).
Isn't that *enormous* for a sensible length of cable?

 

[Atmasphere]

^^You are forgetting that was done with a 3 1/2 digit DVM- its likely the actual voltage drop was a little lower. You also want to include the termination resistance- I think you will find that if a cord is stressed, that the ends are where it heats up before the middle does.

 

[Johnny Vinyl]

I was enjoying this thread, but once again it started turning all technical again. I give up!

 

[Whatmore]

^^You are forgetting that was done with a 3 1/2 digit DVM- its likely the actual voltage drop was a little lower. You also want to include the termination resistance- I think you will find that if a cord is stressed, that the ends are where it heats up before the middle does.

Again, pardon my ignorance but what is a 3 1/2 digit dvm?

 

[Atmasphere]

Sorry- DVM is Digital Volt Meter. '3 1/2 digits' refers to the fact that it can show a number like .005 and has the ability to put a 1 on the screen like this: 1.005

So the 1 can be toggled on and off but a 2 or other number can't be displayed in that position. IOW it is the 'half digit'. Obviously there are resolution problems with a meter like this so if you are inbetween values you really don't know what the actual reading was. This is an area where an analog meter can be helpful- assuming its accurate... Anyway, a 3 1/2 digit DVM is the workhorse minimum ("Good enough for government work"). We have a 6 1/2 digit meter in the shop too but its settling time is noticeably longer (its an older Fluke unit) and usually we don't need the resolution to get the job done.

 

[stehno]

What is it about this seemingly magical number of burn-in hours that seems to be the breaking point for 99% of new gear out there? Its the #1 answer for everything from cables to amps to sources to speakers.

It's probably more like 97%. But IME, the vast majority of components and speakers seem to take between 150 - 200 hours with just a few exceptions. So it would seem 200 hours is a relatively safe number to reference.

 

[microstrip]

Ralph Karsten of Atmasphere advises us to run tubes for at least 72 hours continuously, **filaments only**, no B+ or bias voltages, to cure the cathode coating. I have found that many tubes need between 50-100 hours operation to stabilize, reducing their measured noise after this period.

 

[sbo6]

I can say with 100% confidence that breaking in speakers and xovers matters. When I bought my speakers they imaged like shite and sounded compressed for lack of a better descriptor. Even my wife, without me prompting her stated, "Is that how they will sound, I think I like the older ones better". ~200 hours later (as recommended by the mfr, Usher) night and day difference. Even my wife was blown away and asked what else I had done to improve the sound so much. I had a similar experience when I upgraded the xovers albeit less of a dramatic difference.

 

[Atmasphere]

Ralph Karsten of Atmasphere advises us to run tubes for at least 72 hours continuously, **filaments only**, no B+ or bias voltages, to cure the cathode coating. I have found that many tubes need between 50-100 hours operation to stabilize, reducing their measured noise after this period.

Just for the record, this process is called 'preconditioning' and is not part of the break-in process. It is important for power tubes in general though, and can double the life of the tube if it is done before any B+ is applied.

 

[caesar]

I can say with 100% confidence that breaking in speakers and xovers matters. When I bought my speakers they imaged like shite and sounded compressed for lack of a better descriptor. Even my wife, without me prompting her stated, "Is that how they will sound, I think I like the older ones better". ~200 hours later (as recommended by the mfr, Usher) night and day difference. Even my wife was blown away and asked what else I had done to improve the sound so much. I had a similar experience when I upgraded the xovers albeit less of a dramatic difference.

For guys who have tubes, playing your system for a week non-stop and have it sound like crap is an extra pain