Hi
I need help diagnosing a problem with my ML 333:
I keep the amplifier always on stand-by. Today there was a sound similar to a capacitor discharge and the amplifier went from standby to off. No smoke, no fire. As I was wearing headphones when that happened and I did not use the amplifier at that time I am wondering why such a sound would come out of the speakers. So perhaps it was noise coming out of the amplifier (never experienced that before, though...)? At that moment I was not able to identify the source.
I disconnected the power cord, let the unit sit for one hour, re-connected the power and pushed the power switch. I heard the initial click of the power relay and then the unit went back to off (no blinking power LED at the front panel). Pressing the power button again did not result in any action (no more relay click).
I checked all loudspeaker connections for short etc., no issues. Again: When this happened the 333 was on standby.
The unit was recapped and checked/updated by an authorized ML repair shop approx. one year ago and was working fine until today.
I am quite experienced with Audio amplifier repairs but so far resisted on working on the ML due to the difficulty in obtaining parts. However, as I am not able to spend another $1,000 - $1,500 for another repair I am hoping that someone on this forum can provide me with some guidance on how to diagnose this problem.
I also do not have schematics, although I found some docs on this forum about the 336. Not sure of I can use them here.
Any support is greatly appreciated. Thanks!!!
I need help diagnosing a problem with my ML 333:
I keep the amplifier always on stand-by. Today there was a sound similar to a capacitor discharge and the amplifier went from standby to off. No smoke, no fire. As I was wearing headphones when that happened and I did not use the amplifier at that time I am wondering why such a sound would come out of the speakers. So perhaps it was noise coming out of the amplifier (never experienced that before, though...)? At that moment I was not able to identify the source.
I disconnected the power cord, let the unit sit for one hour, re-connected the power and pushed the power switch. I heard the initial click of the power relay and then the unit went back to off (no blinking power LED at the front panel). Pressing the power button again did not result in any action (no more relay click).
I checked all loudspeaker connections for short etc., no issues. Again: When this happened the 333 was on standby.
The unit was recapped and checked/updated by an authorized ML repair shop approx. one year ago and was working fine until today.
I am quite experienced with Audio amplifier repairs but so far resisted on working on the ML due to the difficulty in obtaining parts. However, as I am not able to spend another $1,000 - $1,500 for another repair I am hoping that someone on this forum can provide me with some guidance on how to diagnose this problem.
I also do not have schematics, although I found some docs on this forum about the 336. Not sure of I can use them here.
Any support is greatly appreciated. Thanks!!!
There was a part change to fix issues like yours. Replace U1 on the voltage gain board with LT1097CN8 OPAMP.
@willcycle: Thank you so much for your response. I will do that component change. Meanwhile: After I pulled the amplifier from its position I noticed quite some fluid underneath, possibly electrolyte. So I am going to dismantle the amp today and check for the source. Would you know the preferred capacitor brand and series for replacements inside an ML?
I would say any first tier brand like Nichicon would work fine. The challenge I believe is finding the correct size.
Thanks. I posted a new thread about capacitor sourcing. Unfortunately it looks like that (at least in the U.S.) the major suppliers only offer the same brand with that rating (50,000uF/125V): Cornell Dubilier. As that brand capacitor failed in my amplifier only after seven months of operation I would prefer a different brand. Even this one is back ordered 14 weeks...
A different cap brand will affect the sound. May be better or worse according to your taste but certainly won't be the same.
Nichicon sells a 1000 hour cap service life cap like everybody else. But if you pay 20-40% premium you can buy 3000, 5000, sometimes higher hours service life. Higher service life gets better sealant and maybe more volume.
digikey US stocks today a nichicon 47000 uf 160 v cap with 5000 hour rating at 85 C . Blocked
1.25" tall 3.54" diameter screw terminals. $156 each.
LNR2C473MSE I guess diyaudio doesn't like people finding parts for other people.
If you read the datasheet CDE has a looser end of life ripple spec than nichicon, rubicon, panasonic, vishay. I in general don't like to buy CDE, partly because they have always come from a country where lying on paperwork was trained by the government.
digikey US stocks today a nichicon 47000 uf 160 v cap with 5000 hour rating at 85 C . Blocked
1.25" tall 3.54" diameter screw terminals. $156 each.
LNR2C473MSE I guess diyaudio doesn't like people finding parts for other people.
If you read the datasheet CDE has a looser end of life ripple spec than nichicon, rubicon, panasonic, vishay. I in general don't like to buy CDE, partly because they have always come from a country where lying on paperwork was trained by the government.
Last edited:
A quality brand capacitor (and that includes CDE) should not be "wetting the bed" after less than a year. If the parts were legit to begin with, I would suspect either a failed rectifier diode (electrolytics don't like being subjected to lots of AC voltage) or substantial overvoltage (most likely as a result of excessive mains voltage... in some areas of the world the grid is that unstable).
BTW, you leave an amplifier rated 350 W @ idle running 24/7? 😱 I get trying to minimize power cycles, but that would be over 3000 kWh a year! (Not counting related extra A/C power usage if you have that.) Enough for an entire 2-person household around these parts. The standby mode drops power consumption to 200 W, which still is orders of magnitude away from modern-day efficiency standards (0.5 W max, though I would be happy with 2 W here).
Since I doubt that there is any master/slave outlet strip that can handle this beast (45 A max), perhaps the ML authorized place has got an idea how to reeingineer it to bring it more in line with the times.
BTW, you leave an amplifier rated 350 W @ idle running 24/7? 😱 I get trying to minimize power cycles, but that would be over 3000 kWh a year! (Not counting related extra A/C power usage if you have that.) Enough for an entire 2-person household around these parts. The standby mode drops power consumption to 200 W, which still is orders of magnitude away from modern-day efficiency standards (0.5 W max, though I would be happy with 2 W here).
Since I doubt that there is any master/slave outlet strip that can handle this beast (45 A max), perhaps the ML authorized place has got an idea how to reeingineer it to bring it more in line with the times.
@indianaojo and LJYS: Thank you very much for your input and sourcing suggestions! I was not aware that CD caps are made in China. And I am wondering why Mark Levinson service would use these caps rather those made from a more reputable source.
I will look into availability.
Would you recommend to replace all 4 caps?
I will look into availability.
Would you recommend to replace all 4 caps?
I checked the above suggested caps with Mouser and Digital-key: The Chemicon is NLA and the Nichicon does not mechanically fit: The max diameter I can fit is 77mm, the Nichicon has 90mm (due to its 160V rating).
There are caps available with a 100v rating (instead of 125). How critical is that if I would decide to install 100V caps instead?
There are caps available with a 100v rating (instead of 125). How critical is that if I would decide to install 100V caps instead?
It is almost as if you got a lower grade black caps in your 333 after revision.
In the image below you can see a 333 with the trusted 5000hr lifetime blue caps.
These blue Cornell Dubilier caps are still available, but not exactly the 50.000u/125V, but I also don't know what the original value was, was it indeed 50.000uF/125V?
http://www.farnell.com/datasheets/621881.pdf
A 45.000uF 160V with 76mm diameter would probably fit and perfectly fulfil it's duty.
Do you know the voltage on those caps, I would be careful to lower the specified voltage.
Hans
.
In the image below you can see a 333 with the trusted 5000hr lifetime blue caps.
These blue Cornell Dubilier caps are still available, but not exactly the 50.000u/125V, but I also don't know what the original value was, was it indeed 50.000uF/125V?
http://www.farnell.com/datasheets/621881.pdf
A 45.000uF 160V with 76mm diameter would probably fit and perfectly fulfil it's duty.
Do you know the voltage on those caps, I would be careful to lower the specified voltage.
Hans
.
Attachments
Hans, thanks! I am currently searching for Cornell Dubilier DCMC503M12SDP2D. I guess this is the one your are referring to. Unfortunately, lead time 14-18 weeks...
I do not know the original value (no schematics) and I do not know the actual voltage at the caps. If all fails I guess I need to wait....
All other brand caps are either rated at 100V or 160V. The 160V would not fit mechanically
I do not know the original value (no schematics) and I do not know the actual voltage at the caps. If all fails I guess I need to wait....
All other brand caps are either rated at 100V or 160V. The 160V would not fit mechanically
You could ask Dimitri on this forum or send him an email at 🙂cop: email removed)
He worked for Harman Kardon and had acces to all ML documentation.
I would be surprised if he could not tell you the original value of those caps.
Hans
P.S. Since one channel is still functional, you could disconnect the other channel and measure the voltage on the big caps.
He worked for Harman Kardon and had acces to all ML documentation.
I would be surprised if he could not tell you the original value of those caps.
Hans
P.S. Since one channel is still functional, you could disconnect the other channel and measure the voltage on the big caps.
Last edited:
You can buy 5 10000 uf snap in caps. I would buy no voltage lower than 125 unless you have the schematic and the rail voltage is 100 or lower. Cut a octagonal board of Nema CE laminate McMaster-Carr
drill holes for the legs. Stack caps cylindrically to fit space. Parallel connect 5 caps with 18 ga wire. Glue to the chassis with banned wall board adhesive (not compliant with California's stupid air board rules). Or some people like hot melt. I've never had success with it.
drill holes for the legs. Stack caps cylindrically to fit space. Parallel connect 5 caps with 18 ga wire. Glue to the chassis with banned wall board adhesive (not compliant with California's stupid air board rules). Or some people like hot melt. I've never had success with it.
So I just found a suitable capacitor at Mouser (DCMC503M125DF2B), with the correct diameter, but slightly shorter (which is ok). I decided to replace all 4, as one is definitely bad and another one looks dubious around the vent hole. Pricey, but I do not want to open this amp again in the near future.Let's hope that the amp will work again after replacing the caps and no other problem was caused by the leaking cap...
Thanks guys!! I just contacted Dimitri for the original values and the voltage at the caps. Although I now have hopefully found a good replacement I think it is good to have that information (and the suggested go-around from indianajo) at hand if need be.
You are a glutton for punishment. Those are CDE 2000 hour rated service life caps. Okay if the amp is going to sit on the shelf. My house amp runs 2000 hours a year. If your rectifier is not shorted as mentioned previously, CDE got you the last time. You need to check the power supply bridge diodes or rectifiers with the diode scale of a DVM. .5 to .75 v forwards, 9999 or ---- backwards.
Last edited: