Stephen M

Thanks Roy, what you've said makes a lot of sense. There is definitely a trap when chasing the measurements rather than using my ears. The overall sound is very good, and I'm really happy with how things are now. I've done a bit more room treatment, and will attach the last measurement. Looks like I need to back off on the high range attenuation now, but things are in a good spot now for fine-tuning.

The 3a Improved doesn't have a mid frequency control, it's a 2-way switch. Do you think the lower output of the midrange is to be expected? The tweeters are new HiVi substitutes, and might have a higher output than the originals, although this is a guess. I don't think either one is faulty, as they both have a very similar response. Another possibility might be that C2, a 20uf Solen MKP that is in the midrange path, needs replacing. This is the only cap in the crossover I haven't replaced, as it tested OK in both speakers.

Hi Roy, There's a technician here in Melbourne who makes voice coils, and the thought crossed my mind that he might be able to do some work on these midrange drivers. Assuming they can be worked on... Stephen

Well, I did quite a few other things when recapping, so it's hard to know what the individual effect of each was. I like the sound now, and sealing the cabinets has tightened up the bottom end a lot. Room treatment will be a big factor in the sound too. I'm just wondering whether it'd be worthwhile getting the mid range serviced.

Here are a couple of full range sweeps with the mic at approx. 1m distance, pointing directly at the speaker at a point between the mid and tweeter. The room is untreated so there are several nulls I need to try to eliminate, but the general response may be seen in white. y

Does anyone else think these measurements indicate a problem with the midrange drivers?

Roy has offered some advice regarding these tweeters. He pointed out that the AR-3a Inproved crossover already has a 0.1mH inductor across the tweeter, and suggested keeping this in place and testing before making any changes. At the moment I'm pretty happy with this setup, but time will tell.

After recapping all but the 20uf and big compulytic 150uf, I grabbed some magnitude spectra, close mic'ing each driver. The crossover points I got were: Left low/mid - 497Hz mid/high - 6147Hz Right low/mid - 432Hz mid/high - 6261Hz I've now replaced the old compulytic 150uf with a mundorf bipolar electrolytic, and sealed the cabinets.The crossovers are now measuring: Left low/mid - 508Hz mid/high - 4787Hz Right low/mid - 506Hz mid/high - 4865Hz Attached some grabs of the second measurements.

Thanks Kent and Roy, At this point I'm not inclined to change the polarity, as I like the sound as-is. Following up on Kent's post, I've done some more reading and have learned that there are indeed instances where the phase of tweeter or midrange are inverted, depending on the crossover. I don't have the knowledge of crossovers or filter circuitry to understand the reasoning behind this fully. I'm sure the authors of this schematic got it right. The HiVi tweeters sound very good to my ears. Initially the system seemed very bright, but it has improved greatly since I put a rug over the bare floorboards between the speakers and my listening position. I intend to continue tuning the room to flatten the response as much as possible, starting with bass traps to address significant nulls at 72Hz and 216Hz. Cheers Stephen

I've also done a quick survey of other schematics posted here. In each case the tweeter is wired with the same polarity as the other drivers. I'm retuning to my initial question: Is it possible that the tweeter should be wired opposite to what is shown in the above schematic?

Hi Mike, Thanks, I guess I should have asked a different question: "Why is the tweeter wired differently to the other drivers? Why is tweeter +ve wired to the -ve side?"

Hi, I'm a bit confused about the wiring of the tweeter in this schematic. It seems to me the +ve terminal of the tweeter should be connected to black, not yellow as shown below. I'm basing this on the fact that the woofer and midrange have -ve to ground, and so it logically follows that the tweeter should too. I'm also assuming the yellow dot on the tweeter terminal indicates +ve. Am I wrong, or is the schematic? I suspect I'm wrong, but anyway.... Cheers Stephen

Hi, Just wondering if anyone has set up their AR-3a's sideways rather than vertical? Interested to know how they are set up horizontally, facing the listener - woofers to the inside? Cheers Stephen

Thanks for the replies. I ended up taking the speaker outside and using the soldering iron to soften the glue (very smoky job!). I then cleaned as much as I could with isopropyl alcohol and did the recap. Cheers Stephen

Hi, I have a general question about removing old hot glue from my AR-3a Improved crossovers. I'm re-capping them, and there is a huge amount covering the switch and terminal posts that I need to remove. I've tried using my soldering iron to soften the glue and then tease out the wires, but I've realised that I need to remove it and clean up the residue so the solder will stick when I install the new caps. Are there any tricks to doing this? I have a heat gun but am hesitant to heat up all the components on the crossover. I've attached a photo with an arrow pointing to the blob in question. Cheers Stephen