I was experiencing "drop-out" and "fade-out" of one channel, and then sometimes the same with the other channel. I tried everything, including reversing cables. I heard the same problem through my headphones, so it certainly was not speaker/speaker cable related. I then tried a power cable with a very tight IEC connection, and it seems to have solved the problem. Has anyone experienced anything similar?
No, and it would be very unlikely that the power cable has "solved" the problem. The incoming AC line feeds the power supply without discriminating between channels. Even with a dual power supply amplifier, the amplifier's single AC input isn't split until after it enters the component. A power supply problem could cause an intermittent channel though I have no idea what "fade out" actually means in terms of operation. Typically a channel either works or does not work. Any loss of power is a sign of problems in the amplifier.
With the symptoms you have described and the bare minimum of information you have provided, I would strongly suggest you have a technician go over the amplifier. This could be a simple problem that can be repaired on the spot or, as Click and Clack say, you might have a problem where you won't be around for a follow up call. Take the amp in to a repair shop. Wiggling or replacing a power cable to restore proper operation is not a "fix" and should be taken as an indication of more serious problems inside the amplifier.
Thanks for your response. Let me be more specfic. The amp is a Creek 5350SE integrated amp with a MM phono gain stage.
After about a half-hour to an hour of use with a Marantz SA11-S1 SACD/CD player, Gallo Reference 3 speakers, and Gallo Reference Sub-Amp, the output of one channel of the Creek amp would sometimes completely drop off. At other times, the channel would not drop off entirely, but the output would be reduced so that the soundstage image would drift to the other channel (a "fade-out" type of effect). Sometimes the problem was with the other channel. When I held the power cable tight against the amp socket, I could hear that weak channel actually become louder.
Finally, I tried three things: First, I replaced the power cord with a cord that provides a very tight IEC grip. Then, I checked the quiescent current (bias) of the amp and adjusted it to Creek specifications (to 21.5mV bias or 65mA of current), then finally I lifted the the three large ribbon cables in the amp (that fit onto pins at the front of the amp) and cleaned them and the pins thoroughly (including blowing jets of air into the female receptacles on the cables themselves). This was after sending the amp to a tech referred to me by Roy Hall (the tech ran the amp for two days and could find nothing wrong, except that he said the amp was running too hot and adjusted the bias down to about 5Mv) and after several conversations with the tech.
That was a couple of days ago, and I've run the amp several times now for about 10 hours total without any problems. It seems completely bizzare to me, and I was wondering whether anyone else has experienced anything similar to this.
I also had the same problem with my 5350SE, and had a hell of a time troubleshooting it. I too thought that it might be one of the ribbon cables inside the amp, so I tried cleaning their connectors and using pro-gold on their contact pins, etc., to no avail.
One day, I decided to run the amp with its cover off, and just waited until I heard the "fade-out/drop-out" effect you describe, and then tried touching some of the components (including the ribbon cables at various points) to see if it was a faulty contact.
I discovered that a short & wide (16-way) ribbon cable from the small circuit board with volume control potentiometer to the main circuit board (the one with all the capacitors) was making a poor connection on the smaller of the two circuit boards. The connection there is made with an IDC (Insulation Displacement) connector whose base is soldered onto the board. For some excellent details on this type of connector, you can download the following PDF file (or do a Google search on "IDC connector"):
I found that reseating the cable onto the IDC connector fixed the problem.
I did the repairs myself: With great care (at the risk of breaking it!) I removed the top cover of the IDC connector. I then lifted the cable from the contact jaws of the IDC connector and trimmed off about 1/4 inch from the end of the cable. Then, with the help of a dull knife, I gradually pushed each of the 16 cables into the contact jaws of the IDC connector, until the entire ribbon cable was fully seated into the connector. Then I replaced the IDC connector cover. The problem went away for good!
Note, there is probably more than one way of fixing this problem, including replacing the cable+ICD connector, or premanently removing the IDC connector entirely and soldering the cable directly to the small circuit board (both approaches require taking some of the amp apart and involve some very careful soldering). But if you're not confident with your electronics repair skills (and your amp is still under warranty), you're better off taking your amp to a repair technician. Just make sure to tell him about what I wrote up here.
I hope this helps.
UPDATE: After about a month, I started having the same problem. Apparently my fix was not "for good" as I originally thought. So I decided to get rid of the IDC connector altogether, and connect the cable directly to the circuit board. (I hear Creek fixed this problem with their connectors in later amp models. Can anyone confirm?)
This was a very difficult job, as I did not have a special desoldering iron for the base of the connector (one of the reasons why this is best left to more experienced technicians), and I had to make sure each of the 16 cables was properly tinned with solder, and then matched to the correct spot on the circuit board. I used Cardas solder for the job.
The "brute force" method seemed to work, and I have not had problems in the past 3 months. I will post an update if anything changes for the worst, in case anyone is interested.