It depends what you’re trying to achieve, what drivers you have, all sorts.
Always easy to criticise though, I enjoy that
It depends what you’re trying to achieve, what drivers you have, all sorts.
Always easy to criticise though, I enjoy that
Are you critising from the Turan, or are you in the pub now?
Tomorrow…
Ah…
Little bit of a fiddle about with time alignment and EQ
Only just started listening to it so no judgement yet.
That bump at 3k was nasty, put a -5db cut there. Sounds way better. Really need to address the crossover point. I set the time alignment by inverting phase on the mid then testing to find the deepest null. When I switched the phase back it looked ok, however measuring from 1 metre I still get a dip and if I change from 48db slopes to 12db it drops about another 6db. Maybe need to redo the invert/null thing with the 12db slopes.
IDK
The dip could be driver response - look at the response of each driver and see if there is a dip.
After that, you need to consider phase. It may be that the drivers are not summing correctly. Drivers change phase over frequency, and if you have frequencies where two drivers subtract rather than add, it doesn’t sound great.
If you correct reach driver to be flat (either using minimum phase FIR filters or standard parametric EQ), then they should have no phase oddities. You can then apply your crossover and you should get perfect addition on the crossover and phase correct speakers.
I did a half assed attempt at that. Measure each speaker at 1 metre, tried to get a flat response (didn’t try very hard as I couldn’t work out how to get REW to do it but now I know how). Then added the crossovers and did the invert/null thing.
Will have another go and try harder at flat responses first.
I tried a bit harder at flat responses measured near field, this is what the whole lot looks like from the listening position. Still need to do time alignment which I’m just going to do with a tape measure.
How is it sounding?
In absolute terms, astonishing. Way better than anything I’ve owned which when you consider how much my last speakers cost is a result.
There are definitely rooms for improvement. Physically lower tweeter, time alignment, better braced bass baffle, better damped woofer chassis etc.
Happy new year. Started to think about trying a folded horn for bass instead of the OB 15"s I have currently. First thing is to see how the current 15s would behave horn loaded.
Happy New Year Ant
Glad to see that your project is progressing
Well to be honest it sounds great and this is really fueled by a desire to move away from active crossovers and have something a little simpler in that regard. I did hook up my TT recently using a TVC pre. It didn’t work right but eventually, I just plugged the phono stage straight into the crossovers and played at full volume. It wasn’t overly loud and sadly sounded a bit shit, so I went back to using the streamer.
A folded horn for bass is very big indeed - my Lab subs are about 4’ x 3’ x 18". They use a smaller driver, but using a 15" won’t make much difference in that respect. Also 1) a horn probably needs a steeper cutoff, so isn’t the way to move from an active XO, and 2) the drivers are unlikely to work in a horn if they work in OB.
If you want a simple system that doesn’t need an active XO, I would strongly recommend the Lampizator P17, which I built and took to Scalford a while back:
http://lampizator.eu/speakers/projects/P17/Endorphine%20from%20Kingston%20Kitchen.html
They are easy to build, the Saba Greencones cost very little and are readily available on eBay Germany, the crossover is simple and they sound great. @Jim almost bought them to replace his Wilsons but he couldn’t see his TV with the size I made them
I do prefer my horn system to them, sure, but it’s a very different thing. Everyone who has a horn system has gone through a world of pain, and the vast majority end up digital active. If you want simpler and analogue, build a pair of those and use a low power amp like an AI series 300 and it’ll sound fantastic.
You’re not rong
Not everyone
*yet.