Hello All,
I have a Creek Classic 5350SE integrated amplifier dating from 2007 that has been in daily use. A few days ago, the motorized input selector switch stripped it’s gears, and is no longer functional. I am considering rehabbing the unit by removing the motorized switch and replacing it with a manual switch. However, I estimate that the ten 2200µF power supply Al electrolytic capacitors (Samwha brand) have over 40,000 hours on them. I have done a little research, and find that is about the service life of such capacitors. Therefore, I am thinking that attempting a rehab is a waste of time and money. Any agreement or disagreement?
Cheers,
ceulrich
I have a Creek Classic 5350SE integrated amplifier dating from 2007 that has been in daily use. A few days ago, the motorized input selector switch stripped it’s gears, and is no longer functional. I am considering rehabbing the unit by removing the motorized switch and replacing it with a manual switch. However, I estimate that the ten 2200µF power supply Al electrolytic capacitors (Samwha brand) have over 40,000 hours on them. I have done a little research, and find that is about the service life of such capacitors. Therefore, I am thinking that attempting a rehab is a waste of time and money. Any agreement or disagreement?
Cheers,
ceulrich
The way to prove or disprove the age of your mains capacitors, is to test the wattage out of the amp. If your speakers are rated higher than the amp rating, you may use the speakers. If not, 4 ohms or 8 ohms of resistors wattage higher than the amp are appropriate. I buy mine at surplus houses. Most recently 1 ohm 100 W $11 each electronicsursplus.com. I mount resistors on steel sheets so they won't burn the carpet or coffee table.
Turn up until amp until it clips, back off a little, measure voltage out on resistors with RMS capable DVM (waste of money IMHO). analog VOM ($14 ones with 20 & 50 vac scales are fine), oscilloscope. Vav = (.707/2)*Vpp of scope.
OTOH, 2200 uf 80 v caps are about $16 for 4 in the 2000 hour 85 C (2-4 years life) grade, $24 in the 5000 hr 105 c grade (8-20 years) $64 in the 15000 hours 85 C grade (rest of your life) . www.newark.com/w/c/passive-components/capacitors/aluminium-electrolytic-capacitors/snap-in-screw-terminal-aluminium-electrolytic-capacitors?capacitance=2200uf&voltage-rating=80v|100v&range=inc-in-stock|not-exc-nic&min_max=f1000360#
You really want to buy a class D amp for $$$ with maybe 6 years life and put this one in the dump? The next one will be surface mount (hard to work on) and the schematics of many are not available (Behringer, Klipsh for example). New amp saves time, wastes dump space and puts copper, tin, lead, in dump for a million years. Europeans, your old electronics are collected, put in a container, shipped to a 3rd world country where they pollute the surface of the land and the rivers in bad neighborhoods. They are burned for the copper. Blown amps with a dozen blown transistors are iffy in payback, but I find geriatric amps that put out low wattage are low hanging fruit. Mains caps, maybe the input caps (bipolar elecrolytics are always low-life, even new ones). I can blast out the neighbor's fireworks on the 4th of July with my 600 w speakers & re-e-capped 1350 W amp. Or play an audience of 600 if I should ever get popular (HA!, my audiences are usually 12 or fewer).
Turn up until amp until it clips, back off a little, measure voltage out on resistors with RMS capable DVM (waste of money IMHO). analog VOM ($14 ones with 20 & 50 vac scales are fine), oscilloscope. Vav = (.707/2)*Vpp of scope.
OTOH, 2200 uf 80 v caps are about $16 for 4 in the 2000 hour 85 C (2-4 years life) grade, $24 in the 5000 hr 105 c grade (8-20 years) $64 in the 15000 hours 85 C grade (rest of your life) . www.newark.com/w/c/passive-components/capacitors/aluminium-electrolytic-capacitors/snap-in-screw-terminal-aluminium-electrolytic-capacitors?capacitance=2200uf&voltage-rating=80v|100v&range=inc-in-stock|not-exc-nic&min_max=f1000360#
You really want to buy a class D amp for $$$ with maybe 6 years life and put this one in the dump? The next one will be surface mount (hard to work on) and the schematics of many are not available (Behringer, Klipsh for example). New amp saves time, wastes dump space and puts copper, tin, lead, in dump for a million years. Europeans, your old electronics are collected, put in a container, shipped to a 3rd world country where they pollute the surface of the land and the rivers in bad neighborhoods. They are burned for the copper. Blown amps with a dozen blown transistors are iffy in payback, but I find geriatric amps that put out low wattage are low hanging fruit. Mains caps, maybe the input caps (bipolar elecrolytics are always low-life, even new ones). I can blast out the neighbor's fireworks on the 4th of July with my 600 w speakers & re-e-capped 1350 W amp. Or play an audience of 600 if I should ever get popular (HA!, my audiences are usually 12 or fewer).
Hello indianajo,
Thanks for the comment. Sounds like you are suggesting I go ahead with the rehab including new PS caps. Good point about a new amp. Let’s see if I get any additional comments in that direction.
Cheers,
ceulrich
Thanks for the comment. Sounds like you are suggesting I go ahead with the rehab including new PS caps. Good point about a new amp. Let’s see if I get any additional comments in that direction.
Cheers,
ceulrich
Instead of punishing your middle-aged (not old) amp with full power, you could measure the AC ripple on the rails to give you an idea of whether there's enough capacitance remaining in those original caps. Much above 50mVAC would be a problem to me, but I don't know what's normal for that model.
But after you check the ripple and confirm below 50mV, then you could run it to clipping.
But after you check the ripple and confirm below 50mV, then you could run it to clipping.
You would need to have the amp deliver power to a load when making this ripple measurement. Ripple may be low with no power output, but it may rise precipitously when delivering power to a load. That would indicate a possible need for reservoir cap replacement, even if idling ripple voltage is acceptably low.
Hello Andersonix, Hello egelings,
I like the idea of conducting an evaluation that does not push the amp into clipping. I will check and report back.
Cheers,
ceulrich
I like the idea of conducting an evaluation that does not push the amp into clipping. I will check and report back.
Cheers,
ceulrich
I prefer evaluating the rail capacitors without taking the cover or knobs off the amp. No problem, no work expended. Clipping is dangerous to speakers, tweeters particularly. Not the amp for brief periods.
If you have an amplifier which you wouldn’t trust to survive a 5 second tone burst at full power, what makes you think its safe to use with expensive speakers?
Over some years in audio I have seen a few wise advices, one them is dont fix what is not broken.
If this input selector still works manually there is no solid reason to replace it IMO.
If this input selector still works manually there is no solid reason to replace it IMO.
You would not need to push the amplifier into clipping at all. Just have it deliver maybe 1/4 full power or a bit more to the load and watch the ripple voltage. You should be able to see the result with only a couple of seconds' run time, and the amp should be able to safely handle that.
I checked the ripple voltage. With an 8-ohm load and no signal the ripple is 35mVrms for the positive rail, and 25mVrms for the negative rail. With an output of 1W, the positive rail is 65mVrms, and the negative rail is 50mVrms.
With an output of 5W, the positive rail is 77mVrms, and the negative rail is 66mVrms. That’s a bit over the suggested 50mV level, but I guess it also indicates that some life is left. The question is, can I expect another 10,000 hours?
FullRangeMan, interesting suggestion of just using it by setting the selector between detents’. That does work, and I thought about using it that way, but the rest of the family was not so interested.
With an output of 5W, the positive rail is 77mVrms, and the negative rail is 66mVrms. That’s a bit over the suggested 50mV level, but I guess it also indicates that some life is left. The question is, can I expect another 10,000 hours?
FullRangeMan, interesting suggestion of just using it by setting the selector between detents’. That does work, and I thought about using it that way, but the rest of the family was not so interested.
I wouldn´t go that far.
In general, when things start going downhill, they roll down faster and faster, you have surpassed a point where everything started to crumble down.
You won´t need to replace them TOMORROW, but bet than in a year performance will be much worse.
Those ripple numbers seem reasonable to me. When the reservoir caps start to go, you get volts p-p of it.
None of the low watts amps I've repaired hummed. 7 watts of 35 on the ST70, 2 watts of 100 watts on the Allen S100's. Sounded fine, just wimpy without peaks.
I'll bet both those rails had only 20mV ripple when new??
Write down the numbers you just measured (short input, open output) and check again in 5000 hours?
Write down the numbers you just measured (short input, open output) and check again in 5000 hours?
You’ll probably forget all about in 5000 hours. That much run time could be a long time.
I just dont know why you don’t do a full power test and find out once and for all. The result would be unambiguous, and you literally would know in 5 seconds. Want to keep guessing, fine, but don’t complain about it if you do. Don’t have a dummy load rated for full power? Don’t need it. Sand cast resistors will take a factor of 4 overload long enough to take the reading.
I just dont know why you don’t do a full power test and find out once and for all. The result would be unambiguous, and you literally would know in 5 seconds. Want to keep guessing, fine, but don’t complain about it if you do. Don’t have a dummy load rated for full power? Don’t need it. Sand cast resistors will take a factor of 4 overload long enough to take the reading.
I measured the power out at clipping with a 4-ohm load: 132W for both left and right channels. I measured them separately since I only have one dummy load.
Clipping was on the negative peak, so I measured the ripple on the negative rail, it was 360mVrms. The amplifier is rated for 180W, one channel into 4-ohms.
Does that change anyone’s opinion on what I should do with this amp?
Clipping was on the negative peak, so I measured the ripple on the negative rail, it was 360mVrms. The amplifier is rated for 180W, one channel into 4-ohms.
Does that change anyone’s opinion on what I should do with this amp?
Was that 360 mV ripple under load? It would be awfully difficult to get only 360 mV of ripple at full power with bad caps - ripple would be far worse. Multiple volts - I’ve seen 20 volts peak to peak, producing good-awful IMD products between the signal and 120 Hz ripple frequency.
If the ripple voltage was under control, and you‘re still short on power maybe it can’t put out full power continuously in the first place. Perhaps it’s only a 1 or 5 millisecond burst rating? Manufacturers have been known to do that. By “continuously” I dont mean indefinitely - just “steady state”.
If the ripple voltage was under control, and you‘re still short on power maybe it can’t put out full power continuously in the first place. Perhaps it’s only a 1 or 5 millisecond burst rating? Manufacturers have been known to do that. By “continuously” I dont mean indefinitely - just “steady state”.
I assume the transformer is the choke point to peak power (since L and R are exactly the same peak power)?
The ripple on both positive and negative rails increases quite linearly (as it should in theory) with current from idle (0.0A) to 1W(0.35A) to 5W(0.79A), although the negative rail holds its ~0.01VAC advantage. Maybe it's normal that the negative rail started with that advantage, but I don't think your numbers up to 5W indicate any trouble.
Find an external switch box, and keep it chugging along!
The ripple on both positive and negative rails increases quite linearly (as it should in theory) with current from idle (0.0A) to 1W(0.35A) to 5W(0.79A), although the negative rail holds its ~0.01VAC advantage. Maybe it's normal that the negative rail started with that advantage, but I don't think your numbers up to 5W indicate any trouble.
Find an external switch box, and keep it chugging along!