I answered your reply in kind -- in other words you asked for it! The reason I asked about why he went with a 1st instead of a 2nd order is because the steeper slope of the latter typically lets you have a lower crossover point which can lead to raising the soundstage (and that's usually a good thing).

If true, then you're correct -- but that's only because I didn't know about either. Is that factory amp cap in series? If so that combo of it and the 4.7uF would have an effect capacitance of 1.40uF while the combo with the 3.3uF would be 1.24uF; so difference in changing that cap (from 3.3 to 4.7) wouldn't make much of an audible difference -- and that doesn't jibe with the real world results discussed here. Can you point me to a thread where they talk about this 2uF cap in the amp?

 

Point was this, in absence of speaker characteristics and a crossover point (neither of which we know) the order of a crossover is essentially meaningless. Sure, each increase in order has a progressively steeper slope which theoretically allows one to lower the crossover point, but that doesn't inherently make any particular order better than any other. With a given tweeter, you can only lower the x-over point to a point beyond which it will break-up. When you said, "the only negative would be...a few dollars", implicit in that statement is that a 1st order is inherently inferior to any higher order, and I would disagree with that.

Yeah, numbers don't lie and it likely wouldn't make much (if any) of an audible difference. And that got me thinking about something when jlukja posted the value. BTW, as for "real world results", what we have is one, single, anecdotal report. That audible difference could very well be the result of a swap for a better quality cap - hell the 3.3 could have been electrolytic and the 4.7 a poly cap, a different manufacturer perhaps, or simply a perception of difference, I don't know.

The question is, why with caps in the amp did El Dude put a cap on the tweet, elevating the crossover point on a speaker with a frequency response range equal to, if not greater than, stock? I suspect that it was a lack of due diligence. Not to dump on the guy as he did post a bunch of useful stuff (in between diatribes), but since he didn't know the HU/amp pathway was balanced, I can only imagine that he didn't open the amp up and take a peak.

As for a thread ref for the caps, there isn't one that I know of, but skepticism is healthy, so you'll have to take your amp, crack her open, and see what's what.

 

But the crossover point at which the SAME tweeter will break up is actually different between a 1st and 2nd order crossover, and that is precisely due to the steeper slope of the 2nd order! That is almost entirely irregardless of the speaker characterisitics. This kind of like determining the vent frequency of an existing ported enclosure. Most people would assume to need to know the specs of the driver to calculate that, but that is false. You can determine the vent frequency of an enclosure totally irregardless speakers specs (because it's only a function of enclosure volume, vent diameter, and vent length).

I wasn't taking one person's (jlukja's) comments as truth, but I was mearly stating they didn't agree with another person's (yours) comments -- although you do bring up some other valid reasons that could lead to an audible difference.

Know of any good threads that discuss the stock amp and its insides? I'll do a search, but I thought you might have good recommendation.

That's just it, I don't have the amp -- because I don't have my TSX yet.

 

Just looking at the capacitors, the one I ordered and the one that came with the Avincar speaker appear to be from the same manufacturer. Same size, same color, same shape, same leads, the only difference being the way the value was stamped on the side (4.7uF vs. 335K). Also keep in mind that the nominal values are +/- 10%.

I am not an audiophile, but the cap switchout made a BIG difference. With the 335K cap I was adjusting the trebble to "-2" because it seemed the tweeters were too bright and piercing. With the 4.7uF caps the trebble is set at "0" and the highs, although still crisp, are more in balance with the door speakers.

 

Yeah, the less steep the slope, the more area under the curve so to speak, the higher the crossover point would likely need be to avoid cone break-up. I'm just saying that one order isn't necessarily better than an other. With the tweeters on the dash, the lower they can play, likely the better, but with the mids in the door and not knowing the characteristics of the speakers, there could be frequency response ("blending") issues that might make a shallow slope might be beneficial. I was under the assumption, initially, that El Dude made his decision based in part on that, now, I'm not so sure...

I don't think there is much discussion on the amp, little interest really. Shit, I'm not all that interested, I just have a tendency to take things apart. I'll try an take some pics and show you what I'm talking about.

 

Good info, but I'm not out of theories yet. What was the lag time between the tweeter install and the cap replacement? Maybe the cap swap took place coincidentally around the time the tweeters broke in, hence less harsh.

 

LOL, Nice try, but no. The time lag was a couple of months at least. And I heard the difference immediately after the swap.

 

Took some pics (and one scan) of the factory amp, highlighting the caps. The camera resolution is poor but it should be clear enough. There is also a daughter board that sits above the motherboard (likely just to keep it compact) but I saw no need to photograph it.

The top of the motherboard:


The best shot I could get showing the cap rating:


The bottom of the motherboard:


I've looked it over and I figure it's clear, but just in case not, the trace on the bottom shows the caps inline with the tweeter + outputs (TWR+, TWL+). Those two caps are the highpass crossovers for the tweeters, unless someone has a better explanation as to their purpose.

 

Give me some time, I'll think of something...

 

Props to you for doing all this investigative and quantitative work. The better we all know what we're dealing with the better we'll be able to mod the sound system to our tastes.

 

All I did was remove four screws. Anyway, I got a new working theory. I think I was looking at this cap issue wrong. A 3.3 uF or a 4.7 uF cap inline with a 2uF cap would only make ~0.5 dB difference which should not be audible, but that's for any particular frequency. I wasn't thinking about the whole range that the tweeter is playing, that would be half a decibel for all frequencies plus a slight crossover point shift down. I don't know for sure, but the summation of all slight difference would likely account for something. I don't think I need to come-up with some esoteric reason as to why you perceived a big difference.

That being said, removal of the supplied cap and utilizing only the factory cap should make for a dramatic difference. It would be interesting to know if the tweeter were 4 ohms or 6 ohms, but either way (even if it were 8 ohms) the shift would be far more noticeable than the swap you already performed. Acoustically better? That could only be determined by experimentation.

Even if I'm wrong and/or made a math error, it is widely assumed that neither the tweeter nor the 6x9 outputs are full range and you can't unfilter a signal. So, by taking the supplied cap out of the signal path, it's not like you'd be sending a full range signal to the tweeter, it's already filtered by the factory amp. I don't have this setup, so I can't test it, and I guess that's where I'll leave it.