Steve:
Martin Logan's are finicky. I have owned a pair for years. There are a couple of remedies. One thing to keep in mind is that Parasound are bit on the harsh side and not really very complimentary to the MLs. I take it you don't really want to change amps--so I'll offer some other less expensive advice. The first thing I would do is try to get a rough idea of what the frequency response in your room is like. This is difficult with ML's because of the panel dispersion. You can use a RadioShack SPL meter or some other C weighted (or even better flat response) meter. Keep in mind the Radioshack is not flat, it's response rolls off at the bottom and top octaves. If you need the correction factors I can get those for you (they were originally published here by A-gon member Sean). At any rate the unit is done about 2 dB below 100Hz. Use Stereophile or other test CD and plot the frequency. You will need to take several measurements and I would do one speaker at a time. When you take multiple measurements start from the listening position and then move the meter about 6 to 10 inches about that area and take an average of maybe 6 readings. This is a pain, but it's necessary because of the panel dispersion. Even then, it's accuracy is somewhat questionable, but I don't know a reasonably simple way without complicating things. (Pink noise is better, but difficult to plot accurately and requires a very good microphone). Okay, now you should have a plot--is it flat--close, or are the high frequencies out of whack compared to everything else. My guess is frequencies above about 4k are up several dB. There are a few things you can do. The first is to get rid of the first order reflection point. You need something that absorbs at the first reflection point on the side walls at approximately ear height. Sonex works well sonically, but not well in terms of WAF (neither do your blankets). The second point of attack is somewhat controversal, but it's always worth a try. Use 1/2 round tube traps behind each panel. This is very tricky, because I'm sure you've been told that the panel needs a reflecting surface behind the speakers. This is true to a degree. What will happen if you do this is you will need to work on placement--most likely the speakers will have to come closer together and the soundstage will get smaller--a price to pay to get rid of the brightness. You will need to experiment with the distance between the panels and the tube trap. Further away will give a better soundstage, but won't tame the high's as much. The third modification is to build your own power cord for the MLs with a resistor in series--you will have to experiment with the value. I used a 20K Pot to find the right value--then it can be measured and replaced so their is not a pot in the power cord. What this does is lower the kV on the panel and reduces it's efficiency. This will increase the life of the panel and will take away some of the brightness. I have done this with ML monolith IIs. But according to ML, the kV on the panel originally was excessive, so the mod was warranted. I'm not sure that is the case with the SL-3s (in fact it probably isn't), so that's why I offer this as a last resort. I hope some of these ideas help in the interium, but at some point you would be happier with different amps.
Martin Logan's are finicky. I have owned a pair for years. There are a couple of remedies. One thing to keep in mind is that Parasound are bit on the harsh side and not really very complimentary to the MLs. I take it you don't really want to change amps--so I'll offer some other less expensive advice. The first thing I would do is try to get a rough idea of what the frequency response in your room is like. This is difficult with ML's because of the panel dispersion. You can use a RadioShack SPL meter or some other C weighted (or even better flat response) meter. Keep in mind the Radioshack is not flat, it's response rolls off at the bottom and top octaves. If you need the correction factors I can get those for you (they were originally published here by A-gon member Sean). At any rate the unit is done about 2 dB below 100Hz. Use Stereophile or other test CD and plot the frequency. You will need to take several measurements and I would do one speaker at a time. When you take multiple measurements start from the listening position and then move the meter about 6 to 10 inches about that area and take an average of maybe 6 readings. This is a pain, but it's necessary because of the panel dispersion. Even then, it's accuracy is somewhat questionable, but I don't know a reasonably simple way without complicating things. (Pink noise is better, but difficult to plot accurately and requires a very good microphone). Okay, now you should have a plot--is it flat--close, or are the high frequencies out of whack compared to everything else. My guess is frequencies above about 4k are up several dB. There are a few things you can do. The first is to get rid of the first order reflection point. You need something that absorbs at the first reflection point on the side walls at approximately ear height. Sonex works well sonically, but not well in terms of WAF (neither do your blankets). The second point of attack is somewhat controversal, but it's always worth a try. Use 1/2 round tube traps behind each panel. This is very tricky, because I'm sure you've been told that the panel needs a reflecting surface behind the speakers. This is true to a degree. What will happen if you do this is you will need to work on placement--most likely the speakers will have to come closer together and the soundstage will get smaller--a price to pay to get rid of the brightness. You will need to experiment with the distance between the panels and the tube trap. Further away will give a better soundstage, but won't tame the high's as much. The third modification is to build your own power cord for the MLs with a resistor in series--you will have to experiment with the value. I used a 20K Pot to find the right value--then it can be measured and replaced so their is not a pot in the power cord. What this does is lower the kV on the panel and reduces it's efficiency. This will increase the life of the panel and will take away some of the brightness. I have done this with ML monolith IIs. But according to ML, the kV on the panel originally was excessive, so the mod was warranted. I'm not sure that is the case with the SL-3s (in fact it probably isn't), so that's why I offer this as a last resort. I hope some of these ideas help in the interium, but at some point you would be happier with different amps.