Odyssey Amps

[jessearias]

I think it can all be agreed to that electrical components  (wires, amps etc.) need some "burning in" to get them seasoned and operating correctly.

My question is the prolonged burn in period required for Odyssey amps of 500 hours or more. It seems like a lot of hours to get something working right. I can understand a couple of hundred hours, but this seems a bit much.

Plus the fact that they want the amp left on strikes me as kinda odd as well.     Is this a flaw in the design?  I have never left an amp running continuously after it has been broken in. I am curious because their amps have gotten good reviews, but this seems weird.

Just curious as to what other members may think.

[tmazz]

I don't know what type of caps they use in the Odyssey amps, but 500 hours of break-in is not unheard of for certain caps. SO saying this am-p requires 500 hours to completely settle in is not something that would strike me as out of bounds.

As for keeping the amp continuously powered on, I had some Audio Alchemy D/A gear back in the 90s that sounded dreadful for quite a few hours after it power up, even if it was of for as little as five minutes, so I just kept that powered on all the time. My tube gear always got powered down when not in use to conserve tube life, but being completely SS, there was no compelling reason to power down the AA gear, so I kept that powered up for years at a time.

Looking at the specs for the Odyssey, it only draws 30 watts at idle, so it is not going to have a big impact on your electric bill So if they think it will sound better left on all the time I personally would not think twice about going with that.

[Emil]

Tom

Does the amp need to have a signal passing thru it or just being "on" would suffice?

[richidoo]

Output transistors' current bias is affected by their temperature and vice versa. Without temperature compensation scheme the transistors will not receive their full current bias until they reach full temperature. You can't provide higher current when cold or they will have too much when hot, so the heating up process takes longer. The bias affects the SQ because it determines how much zero crossing distortion you hear. A class AB amp will sound less good without its intended amount of bias current.

Reduction of zero crossing distortion is the real answer because that lessens the need for class A bias to sound good, and lessens the time to warm up to operating temp.

A temperature-compensation circuit would assure the output devices are always biased to the correct current no matter hot or cold, so you get the same SQ. Without active temperature feedback bias control, cooling off for just a few minutes takes a long time to heat back up again and the SQ suffers whenever it's below the full bias current.

I don't know if Odysseys have temperature compensated bias current or not.  You can check by looking at the output transistors where they mount to the heat sink. There should be a smaller temp sensing transistor placed very close to, or touching the heat sink near the transistors, usually one sensor per pair of output devices.  Of course there could be other schemes to compensate bias. But I think it's fairly important to have some active method to control bias to assure the amp sounds good very soon after power-on.

[tmazz]

Tom

Does the amp need to have a signal passing thru it or just being "on" would suffice?

Don't know about the amps, but the AA DACs just needed to be powered up. I would shut everything off except the DAC.

[tmazz]

Output transistors' current bias is affected by their temperature and vice versa. Without temperature compensation scheme the transistors will not receive their full current bias until they reach full temperature. You can't provide higher current when cold or they will have too much when hot, so the heating up process takes longer. The bias affects the SQ because it determines how much zero crossing distortion you hear. A class AB amp will sound less good without its intended amount of bias current.

Reduction of zero crossing distortion is the real answer because that lessens the need for class A bias to sound good, and lessens the time to warm up to operating temp.

A temperature-compensation circuit would assure the output devices are always biased to the correct current no matter hot or cold, so you get the same SQ. Without active temperature feedback bias control, cooling off for just a few minutes takes a long time to heat back up again and the SQ suffers whenever it's below the full bias current.

I don't know if Odysseys have temperature compensated bias current or not.  You can check by looking at the output transistors where they mount to the heat sink. There should be a smaller temp sensing transistor placed very close to, or touching the heat sink near the transistors, usually one sensor per pair of output devices.  Of course there could be other schemes to compensate bias. But I think it's fairly important to have some active method to control bias to assure the amp sounds good very soon after power-on.

That makes sense in terms of my experience with the DAC. At the time the Audio Alchemy line was considered a budget DAC that sounded way above its cost. Given that I am pretty sure that they did not allocate much if any of the parts budget to  active temp compensation circuitry

[jessearias]

Thanks Richidoo for the great explanation. I am guessing that the Odyssey amps do not have a temperature compensated bias current. Otherwise, they would not tell you to leave them on all the time.

The Parasound designs must use a temperature compensated bias current because it gets cruising fairly quickly (about 1/2 hour). They also do not make any suggestions as to leaving the amp on continuously.

Which is better (temp controlled, non temp controlled), is debatable. To me, leaving the amp on all the time would shorten the life of the electronics in the long run.

Thanks for the great input from all. 

[Nick B]

I wouldn't know about the technical issues mentioned, but my Antelope Gold dac took over 500 hours to break in. The worst of any component I've ever had. But it's sounding very, very nice right now. I do leave it on all the time

[richidoo]

There are a lot of expensive gears that say leave it on. I personally experienced the SQ difference of cold and hot Schitt Yggydrasil this week. 36 hour warm up made a huge difference. They say leave it on all the time, and it runs very warm for a DAC.

But you never know what's going on inside unless you look, or ask. I'm sure Klaus would be happy to talk about his amp designs with a curious owner. He is a very nice guy.

There's many ways to skin a cat. Amp design is a deep dark art. Lots of different ways up the mountain.

[richidoo]

The theory on the slow break in is related to the use of teflon and other strong dielectric films in caps. Strong dielectrics form static fields very slowly compared to other dielectrics like polyprop and polyester. Until the fields are fully formed, the SQ of the cap will keep changing better and worse. Cables have break in time for this same reason, the static charge fields need time to form in the insulation dielectric.

[_Scotty_]

Another point, if you can't hear the difference in the sound of amp between when the amp is first turned on vs. on for several days then you might as well turn it off when not listening unless it will adversely effect the life of the amp due to power cycling and turn-on transients.
 I started leaving my gear powered up 24/7 over 20 years ago with no effect on the life of the equipment. I only turn mine off and unplug when going out of town.
As far dielectrics are concerned, when voltage is applied to an insulator the electric field deforms the charge distribution in the insulator over time and apparently we can hear this.
Link to pdf doc article discussing Dielectric Insulation & High Voltage Issues by Davide Tommasini, CERN
https://cas.web.cern.ch/cas/Belgium-2009/Lectures/PDFs/Tommasini.pdf
Good basic information.
Scotty