Electronics that must stay "on"

[DonH50]

As I said, "poor" is a relative term, and audio amps are not flight hardware nor have to be space-qualified. Clearly the biasing of the outputs was such that it took a long time to reach steady-state, and it was designed for optimum operation at a bias achieved after some time had passed. Is it the way I would design it? No. But, the designer must have had some reason. BTW, it also seems to me the bias level may have been fairly low if it took so long to heat up the heat sinks.

Aside: One of the issues causing my ARC SP3a1's long warm-up time was the string of zener diodes used in the power supply. Even though they ran very hot, it took a while for the power supply to stabilize, despite it being a regulated design.

As a high-speed designer, bias circuits are not my specialty. I had to design (or leverage) some good ones because a jet pilot or spacecraft (satellite or space shuttle, have chips in both) does not have hours or days (often not even seconds) for the circuit to reach spec. It seems like the bias circuit design gets short shrift in a lot of designs; there's more to it than a lot of caps. And, the rest of the circuit has to be designed to be thermally insensitive as well, a non-trivial exercise. There are trades to be made among cost, complexity, noise, stability etc. and in retrospect far be it for me to question an audio design engineer's design. It is not mission-critical so perhaps the cost savings go into a better signal path.

 

[microstrip]

Most electronic audio equipment sounds better after a warmup time that can go from 10-20 minutes to several hours. It is very easy to criticize the designers for not being able to get the best performance almost instantaneously, but the point is that, most of the time, we are still not able to know from the measurements why there is such difference in sound quality after one minute and after half an hour.

Even with solid state I have found that most units take more than five minutes to get their best, unless they have a stand by function. Preamplifiers are usually also very sensitive to warmup

People are usually prepared to accept that tube equipment must have large warmup times because the tubes take time until the transconductance gets to the proper value. Wrong assumption, as after two minutes the tube parameters are much more stable than the variation due to just changing the tube for another sample of the same tube. The real reason for most changes in sound are due to thermal effects in passive components, such as capacitors and output transformers - something much harder to accept. All IMHO

 

[flez007]

Even with solid state I have found that most units take more than five minutes to get their best, unless they have a stand by function....

I recall all my Gryphon amps needed an hour or more to sound correctly, on the other hand my current EAR amp (tubed) needs 15 minutes at the most to start singing....go figure!

This thread deserves a poll, just to learn where we are standing against the average time.

 

[Bill Hart]

My Lamm ML2 amps take 45 minutes to start sounding like they should. You can count on it like clockwork. If I listen to them while they warm up, starting say, after about 15 minutes, they sound OK, but then at around that 45 minute mark, no matter what the program material, they just start to bloom. the image opens up, the soundstage gets deep and the tonality is much more palpable. I assume this has nothing to do with any of the other components upstream, since the line stage stays on all the time, and the same phenomenon occurs without regard to which phonostage I am using. (Although I know the cartridge needs to get exercised too, I usually consider two LP sides enough for that).

 

[DonH50]

Most electronic audio equipment sounds better after a warmup time that can go from 10-20 minutes to several hours. It is very easy to criticize the designers for not being able to get the best performance almost instantaneously, but the point is that, most of the time, we are still not able to know from the measurements why there is such difference in sound quality after one minute and after half an hour.

Even with solid state I have found that most units take more than five minutes to get their best, unless they have a stand by function. Preamplifiers are usually also very sensitive to warmup

People are usually prepared to accept that tube equipment must have large warmup times because the tubes take time until the transconductance gets to the proper value. Wrong assumption, as after two minutes the tube parameters are much more stable than the variation due to just changing the tube for another sample of the same tube. The real reason for most changes in sound are due to thermal effects in passive components, such as capacitors and output transformers - something much harder to accept. All IMHO

Hmmm... The components I have always found most temperature-sensitive are the resistors, then big (electrolytic) caps, coupling caps (depending upon the type), and the transformer was out of the picture except for large signals (since heat can impact the H curve). One excellent point from your comments is that temperature-stable passive components can be very expensive.

One other tidbit is that, while being the most stable, metal-film resistors are not the quietest resistor type so an SNR/stability trade can be made in favor of less stable but less noisy resistors.

Rats, this whole thread is making me re-think my stance on warmup... Though I have never really argued about it in audio components. I still have no good explanation why a cable would require hours to "break in", however.

 

[bblue]

Rats, this whole thread is making me re-think my stance on warmup... Though I have never really argued about it in audio components. I still have no good explanation why a cable would require hours to "break in", however.

I've had probably two dozen different amps and related gear over the years and have NEVER found a single piece of gear that sounded 'right' when first turned on. Some may take just 15 minutes (rare) to get there, while power amps could take up to two hours or more in some cases. Most preamps don't even have power switches. They need to be left on to sound great on command.

Same for Pro gear in studios. Don't ever walk into a studio to record only to find the staff turning on the console, outboard equipment, tape machines, etc. Some of it sounds absolutely horrid for the first couple of hours. Tubes or solid state -- doesn't matter except for warming cycle. It all needs time.

From the Pass Labs X350.5 amp manual:

"People are often interested in how long it takes for these amplifiers to break in. Perhaps a more salient question is how long does it take them to reach their sonic best after turn on. It takes about an hour for them to fully warm up, and this is where we adjust them first. Then we adjust them again and again over a couple of days, keeping the bias and offset in the sweet spot. Our environment is about 23 degrees Centigrade and the heat sinks will rise approximately 25 degrees C. above that, for an average on heat sink temperature of 48 degrees C. Sonically they are at their best when the heatsinks are approximately 50 degrees C."​

--Bill

 

[Bruce B]

Don't ever walk into a studio to record only to find the staff turning on the console, outboard equipment, tape machines, etc. Some of it sounds absolutely horrid for the first couple of hours. Tubes or solid state -- doesn't matter except for warming cycle. It all needs time.
--Bill

That's for sure.... My console has only been turned off twice in 5 years!

 

[MylesBAstor]

My Lamm ML2 amps take 45 minutes to start sounding like they should. You can count on it like clockwork. If I listen to them while they warm up, starting say, after about 15 minutes, they sound OK, but then at around that 45 minute mark, no matter what the program material, they just start to bloom. the image opens up, the soundstage gets deep and the tonality is much more palpable. I assume this has nothing to do with any of the other components upstream, since the line stage stays on all the time, and the same phenomenon occurs without regard to which phonostage I am using. (Although I know the cartridge needs to get exercised too, I usually consider two LP sides enough for that).

Amps that use interstage coupling like Audionote take hours to warm up. Also remember also the older VAC Rennaissance amps needing longer than usual to stabilize.

Digital gear should never be turned off.

Most other equipment takes 30-60 mins to sound their best after turn on. I find that cj takes a good 30 mins.

 

[mep]

I know that when the alarm clock ‘turns me on’ in the morning that it takes 30 minutes and two cups of coffee before I stabilize.

 

[ack]

Most of my electronics are always on: analog section of CD transport, tuner, phono, preamp, analog section of DAC. The amps need an hour, the R2R not sure, the sub about 30 mins (it's old school A/B), the MLs should really be charged for at least 30 mins. You should see the ceremony I go through to warm up my espresso machine, which includes two blank brews

 

[mep]

My turn on ritual is pretty simple. Neither my Krell KPE Reference phono stage or my Krell KBL have off/on switches and they are intended to stay powered on 24/7 which is what I do. I also leave my laptop server and my D/A converter powered on 24/7. I usually go downstairs 2 hours before I plan on listening and turn on the KSA-250 and my tape deck and the pair of Ampex 350s. I have Foobar set to 'random' and I select my jazz file and I let the system warm up for 2 hours with jazz playing through it. The KSA-250 has lots of square inches of heat sinks that takes awhile to heat up before this amp comes up to full operating temperature. Does it take 2 hours to warm up? No. Do I like it better when the system has been 'cooking' for two hours with music playing through it? Yes, I do.

 

[Btselect]

Is it the heat that determines performance or is it the time needed for the circuit to stabilize and the heat is just the by-product of time on? If it's the latter than I would want to cool the component down to prolong parts life, especially electrolytic caps which have a definite lifespan based on temperature and time. If though it's based on heat then a 105 degree cap would be a better choice than a 85 degree one. Do you monitor time on to change out these parts as well as the transistors that drift with time to maintain optimum performance.

 

[mep]

Is it the heat that determines performance or is it the time needed for the circuit to stabilize and the heat is just the by-product of time on? If it's the latter than I would want to cool the component down to prolong parts life, especially electrolytic caps which have a definite lifespan based on temperature and time. If though it's based on heat then a 105 degree cap would be a better choice than a 85 degree one. Do you monitor time on to change out these parts as well as the transistors that drift with time to maintain optimum performance.

I don’t know who you are directing your comments to. I just paid to have a ‘recap’ job done on both my KBL and KSA-250. I’m not going to worry about logging hours and worrying about when my caps will need to be replaced again. Electrolytic caps are normally good for at least 20 years and today’s caps are better made than caps from yesteryear.

 

[DonH50]

All components change with temperature and current, they are fundamentally related. You cannot really separate them; you can design circuits to be temperature-insensitive but that does not mean no changes with temperature but rather that everything tracks so temperature does not change things like gain etc. The devices change with temperature no matter what, it's in the physics. The time required is a function of temperature and bias design. So the answer to your questions is "yes", sort-of... It's complicated.

 

[Btselect]

All components change with temperature and current, they are fundamentally related. You cannot really separate them; you can design circuits to be temperature-insensitive but that does not mean no changes with temperature but rather that everything tracks so temperature does not change things like gain etc. The devices change with temperature no matter what, it's in the physics. The time required is a function of temperature and bias design. So the answer to your questions is "yes", sort-of... It's complicated.

So I take it that first you should check with the engineer of that unit to see what is the best operating temperature, if it doesn't matter then running fans to cool the unit down to prolong parts life would be ideal, if there is a certain optimum temperature one can hook up fans with a temperature sensor to come on when temperature exceeds the optimum.

The reason I use fans or recommend fans and use higher temperature caps with higher performance is that they definitely prolong the life and performance of your components, I have replaced countless number of capacitors in people's equipment new and old to newer and improved ones, when you can A/B the same two equipment it's much easier to hear the difference, if any, to worn caps or parts and newer and improved parts, at times it's hard to determine changes with just unit, most people would say if it ain't broken don't mess with it, it's only when you hear the difference a newer and improved part makes, is when you want it now.

 

[DonH50]

I would just read the manufacturer's comments, email or call, or just plug it in, turn it on, and see if it changes over time. I have to assume most audiophile components are meant to operate with adequate space around them (per most owner's manuals) in a typical home environment. I doubt you'll have much luck in finding out the optimum temperature from the designer of most audio components but it could be worth trying to contact the manufacturer if you have specific gear and concerns. The few times I have done that and gotten a reasonable response, it has been along the lines of "we burn it in in a room-temperature environment for a day then adjust the bias for optimum performance/soiund".

There is all kinds of documentation about how heat degrades components. As for capacitors, electrolytics in particular vary significantly in value over their lifetime due to heat and other issues.

A number of products use fans with temperature sensors to first turn on then adjust the speed of the fan(s) with temperature. PCs and graphics cards come to mind; some AVRs and power amps, etc. Commercial components are typically rated 0 to 70 degC, industrial -40 to +85 degC, military -55 to +125 degC ambient. I would guess many high-end components use at least industrial-quality components. Many products may use high-grade components in key places and less (cheaper) components where it does not matter (to them).

 

[treitz3]

Well, the Oddysey amplifier had a power switch but it was very inconveniently located in the back. Klauss even mentioned several times in his manual that it was designed to be kept on 24/7. I dreaded unplugging it or even turning it off for more than two or three minutes. The Musical Fidelity AC3r Dual Mono amplifier I have now? I turn it on and she's good to go from the get-go. No more waiting. The longest wait I have right now is waiting about two minutes for the Dodd MLP tubed pre-amp's relay to kick on [safety feature as to not introduce a DC pulse]. 2 minutes I can deal with. 3 days? Not so much.

Tom

Well, you learn something everyday. Please allow me to amend what I have stated here because I was apparently incorrect. Because of the almost daily thunderstorms we have here in Charlotte, I always unplug my system every night in fear that I may not arrive home before the storms hit the house. Well, last night, I goofed up and completely forgot to unplug the system. Not only that, I forgot to turn it off. I came home today, discovered my "oops" and plopped in some tunes and then very quickly realized that it sounded better. It went from extremely pleasing to stellar. I have listened to music non-stop for the past couple of hours and apparently I was incorrect in the assumption that it was good to go from the start.

It's almost as if I squeezed out some of the final 5% and I'm only looking at squeezing out only the final 3% now. That may not be a good way to explain it but it's the best I can come up with right now. The sound stage and imaging didn't change but the micro details definitely came into fruition and I couldn't help but get goose bumps throughout most of the songs and albums that I have played. It's not that goose bumps weren't common, just not quite as frequently as what I am experiencing today. In a word, it's tight. I could kick myself in the rear end for forgetting to turn everything off and unplug the system but at the same time, I feel like giving myself a pat on the back for being so daggum stupid as to leave the system on with the constant threat of lightning always looming. It sounds stellar right now. There's actually a storm within close range as I type and though it's not a threat at this given moment, had I not been home and had Mother Nature shifted said storm my way? It could have yielded the same result I had a couple of years ago when lightning struck my rig. Listening to lightning fry the internal components while the frying was being broadcast over the speakers is my idea of an audiophiles worst nightmare. NOT a sound I wish to ever hear again.

Now I have no idea what component is sounding better or if it's even just one component. It could be the power supply, the pre, the amp, whatever. It could be everything. I just don't know. What I do know is that at this given moment in time, I wish I lived somewhere else where lightning never struck. This sounds superb!!!

Tom

 

[Bill Hart]

We've been getting some severe electrical storms here in the NE and i am constantly going through that same ritual. The only component i normally leave on all the time, as i may have already mentioned, is my line stage, which Vlad Lamm recommends, not just for sonics (although it does make a difference), but also for tube stress. (The tubes are in the power supply only).
It is a little harrowing to worry about your hi-fi gear during these electrical storms- sorta neurotic, but, for good reason.

 

[rockitman]

I keep my main pre and phono pre on 24/7. My amps are not meant to stay on 24/7 as they are class A bias and consume 300 watts at idle. Pass has a standby switch which keeps the cap banks charged while consuming few watts. It also extends the life of the caps remaining on and not going through repetitive warmup/cool down cycles that happen when the amp is turned off completely.

 

[treitz3]

It is a little harrowing to worry about your hi-fi gear during these electrical storms- sorta neurotic, but, for good reason.

I know what you mean. The last strike that hit cost me two amplifiers that I had just gotten back from servicing to bring back to spec. only two weeks prior to being blown up and this was with surge protection. They were a total loss. That right there cost me a pretty penny or two. I guess that was Mother Nature's wedding present for me. Grrrr.

Tom