Parasound A51 - Different Temp Heatsinks?

I too have a Parasound A51 which I originally was running with a new pair of tower speakers which after a couple of hours were causing one of the left side channels of the amp to shut down from overheating protection.  I am using two channels to the left and two to the right for bi-amping the speakers.  These speakers were 8 ohm and this amp went into overheat protection several times over the past few months.  I switched the speakers around to see if the over-temperature protection would remain on the left channels or if it would follow the speaker.  It followed the speaker, so I knew that I had a speaker problem.  The speaker was replaced by the manufacture under warranty for a new one.  

I sold those speakers at a deep discount and recently purchased some nearly new Bryston Model T's which are 4 ohm speakers and hence the amp is generating 400 watts per channel vs. 250 watts with the 8 ohm previous speakers.  Putting out 400 watts should cause the heat sinks to run much warmer as it's delivering a lot more current to the speakers now.  I find that the amp is running on average, quite a bit cooler after reading the temperature with a heat reading infrared gun.  With the 8 ohm speakers, it would run up around 135 to 140 degrees with moderately loud volume.  With the Model T Brystons and 4 ohms, it averages around 115 degrees under similarly loud playing.  However I did notice that the right side of the amp runs about 5 to 10 degrees warmer than the left after prolonged playing.  And last night the temperature did hit 145 degree on the right with the left at 130 degrees after 6 hours of heavy loading.   It did not go into over temp protection at any time, but I don't know what the threshold is for that to happen.  I also don't know if the output devices are on one side more so than the other with other devices such as in the power supply or input sections laid out on certain sections of the heat sinks, so I cannot tell if the temperature readings are due to variation in the output devices or from other influences.

I plan on switching the Bryston's from one side to the other just to see what the readings show.  If there is no change, then I'll suspect that either there are variances in the amplifier such as biasing adjustments, not exact component matching or placement of devices on the heat sinks.  If there is a change in readings, then there is a difference in the speaker loads presented to the amp.  It should prove to be interesting to see what it shows.

 

Hello All!

Thanks for all of your input; just spoke to Tony at Parasound and he stated it was within normal operating procedure. Parasound has outstanding customer service and I am very happy to own one of their pieces in my system.

Also, Al, thank you for your incredible detail in your response. This site has become an exceptional resource for me while I delve into this new hobby. Thanks again!

Richard

Stan, thanks. Your comments about the overall impedance of series and parallel connected speakers are essentially correct, although as you realize this is a 5-channel amp, with the speakers undoubtedly connected to separate channels. The reference to RCAs was undoubtedly a reference to the connections from the preamp to the amp (which has both RCA and XLR inputs).

To be completely precise, three 8 ohm speakers in parallel would result in a combined impedance of 8/3 = 2.67 ohms. The general formula, that works for any number of paralleled speakers and regardless of whether or not they have identical impedances, is that the combined impedance equals the reciprocal of the sum of the reciprocals of each of the impedances. So paralleling an 8 ohm speaker with a 6 ohm speaker and a 4 ohm speaker, for example, would result in:

1/(1/8 + 1/6 + 1/4) = 1.85 ohms

Returning to the original question, I suspect that RW is correct, and there is no problem. Photos at the Parasound site show that the speaker connectors and the AC input connector are all located near the left end of the rear panel (as viewed from the front), while the input connectors are located near the right end of the rear panel. So my guess is that the output transistors and related circuitry, which generate large amounts of heat, are located adjacent to the heat sink on the left side, while low level input and other circuitry is located on the right side. Or perhaps the output stages for 3 channels are on the left, with just 2 being on the right. And very conceivably the power supply is located at the rear center, causing the rear heat sink to become hot.

Also, it is indicated that the amp operates Class A up to significantly higher than average power levels, before changing to Class AB. And keep in mind that it provides 5 channels of high powered amplification (250W into 8 ohms, 400W into 4 ohms, for each of the 5 channels). All of that means a lot of heat. Finally, the design includes over-temperature protection mechanisms, and in the event of an over-temperature condition no music will be heard and both the power switch and a dedicated high temp indicator light will glow red.

No problem, IMO.

Regards,
-- Al

If you have different heat sink temps left to right, your bias currents may be mis-adjusted. A high bias current can make the heat sinks warmer. The A51 may have bias adjustment pots to alter the bias currents.

I would have a good shop or Parasound check the bias current levels. This can be done by measuring the DC voltage across the emmitter resistors on the output power transistors.

Excuse me, I thought you were running them all from the same taps. I was conflating your amp with the A21 my friend has; I didn't understand the RCA connection for the speakers; RCA jacks carry signal usually. The fact that one heat sink gets hotter is troubling. Switch both the signal leads and outputs to the other side and see if that side now gets hotter; if it does suspect connection problems, if not have the amp checked. Does the amp also have binding posts?

If you run 2 pairs of 8 ohm speakers in parallel you get a 4 ohm load; not 8 ohms. Adding a third pair will lower it still further; you may have a 2 ohm load on your amp. I will defer to experts like Al on exact loading but in general when you parallel speakers the loading is half their common impedance in my understanding. In series it is double their impedance if I remember correctly. There should be a lot on here or the net in general on this. I would look it up for you but am in the process of moving.

I would first try to determine if you even have a problem. Its possible that the different heatsinks on your amp are colling different components that operate at different temperatures. I would just call Parasound and ask them. Also, I don't see why your Kef speakers should be too much for your amp. It's probably a very good match.

Hi Stanwal

All speakers are rated at 8ohm and should be easy to drive. I'm wondering if the speakers are so efficient that the amp is putting out only class A watts since it's at normal listening volume (first 10-15 watts at class A then it goes into Class A/B)? If it is a bias issue, what would I need to do to correct it?

Have you considered what the combined impedance of the speakers amount to? It may constitute a very low impedance load; a transistor amp will attempt to double its output when the impedance is halved.For example my CJ 350 barely gets warm driving my 8 ohm Spendors but gets quite hot driving the GamuT
L5s which go down to 2 ohms. Or the bias could be turned too high or it could be normal.

I guess my main concern with the heat would be the eventual breakdown of the components inside. Is that a probability?

Don't sweat it, your amps obviously aren't...

-RW-