Hi Denzil,
"I would appreciate all your feedback on the crossovers I'm using. And would it be possible or necessary to change any components to improve them in any way ?"
It's certainly possible to change the components but 'necessary' depends on you.
Fact: Those ubiquitous white box-shaped resistors are the cheapest you can buy, as far as I know, and it can be heard.
I can hear them degrade the sound . Those who can't hear it are lucky. These horrid things have no place in power supplies either.
Now for the onslaught of: But if the resistance measures the same then it must sound the same, ad nauseum.
For the
measurementalists here:
You are trying to measure something that you respond to at an emotional level!
Fact: The coils are not optimised in their relation to each other. This phenomenon can be heard and measured as I explained earlier.
Your XO shown has a different cap setup. The standard MG15 has, in the HP section, a 6.8uf, then a 1.5uf as part of a shaping circuit and then a 3.3uf + inductor as a notch filter. What they appear to have done, looking at the board and following logically from input for sensible layout is to use a 4.7uf as start of the HP filter. The caps after that are correct. Then, and as I can't see the traces, would imagine that the 2X6.8uf are in parallel giving 13.6uf, for the LP section where 15 or 16uf is usually used.
These different values will alter the response. The biggest change will be the 4.7uf where it should be 6.8uf. This will shift the HP upwards to a higher fequency causing an underlap at the intended XO point resulting in a dip.
Also, what are the coil values? Looking at what has been done with the caps, who knows?
I would unsolder the smaller of the 2 coils with slugs and lie it on its side so it will be like a wheel, able to roll around. The slug must lie parallel to the 4.7uf cap pictured above it.
Solder a 2.2uf cap across the same 4.7uf giving 6.9uf. With allowable tolerances it will be close enough. If you like what you hear then the next step is to replace the parallel pair with a single higher quality 6.8uf.
This of course is based on the presupposition that their configuration is as I mentioned. You need to flip the board over and trace out the circuit. If it is as I think then go ahead, if not still change the orientation of the 1 coil and replace noisy resistors with at a minimum, Mills 12W
The unknown coil values are a problem. With this in mind, do this first: correctly reposition the coil and replace all resistors and report back. The least expensive option is finding the tweeter level you prefer and ignoring the ones not being used.
Tannoy allows selection of 2 resistors to provide 3 settings. Better is what I will do with my next rebuild is to use a 50 ohm L-pad to finely vary the level and then measure the series and parallel legs and then replace the L-pad with fixed resistors of those values.
In a previous response I linked to a different thread about much the same thing as discussed here. It is all valuable info concerning your quest. Here it is again. Please read it again even if you have as the information may make more sense now and there is a pic of how to position the coil. Read on a few pages. I explain why fine harmonic retrieval is so vitally important to realistic timbre and soundstage.
https://theartofsound.net/forum/show...efurb-s/page12
Now it's all about the music, right? As I see you are from Goa. Listen to Ottmar Liebert's, The Complete Sante Fe Sessions, the last track: Morning Arrival in Goa.