🙂System7 wrote :
System7 can evaluate the sound seeing the crossover
Sorry to say and repeat, there is NOTHING wrong with this speakers, speakers that I own.
Some interresting points have been shown concerning those speaker in this thread, in particular for the HDS tweeter.
The last was :
I totaly agree with that, in the past listening room i had, i found the HDS tweeter a little harsh, not in the actual 🙂
The last one was bigger (7mx7m) and with wall not covered and stone floor.
The actual room (5mx5m) has walls partially covered with curtains and a wooden floor covered with carpets.
System7 can evaluate the sound seeing the crossover
Sorry to say and repeat, there is NOTHING wrong with this speakers, speakers that I own.
Some interresting points have been shown concerning those speaker in this thread, in particular for the HDS tweeter.
The last was :
I totaly agree with that, in the past listening room i had, i found the HDS tweeter a little harsh, not in the actual 🙂
The last one was bigger (7mx7m) and with wall not covered and stone floor.
The actual room (5mx5m) has walls partially covered with curtains and a wooden floor covered with carpets.
Last edited:
system7 has given some very good advice so you may want to go over it again. I believe it may be the 830875 and/or it's implementation in the crossover and not the tweeter.
I've designed and built several loudspeakers using those drivers but not both in the same design. Even so I am aware of their behaviour.
The 830875 has varied quite a bit over the years as it went from Danish to Chinese manufacture. Even the Chinese built drivers have had a few changes which could effect sonics from batch to batch, so you may be hearing a variation to the driver that was used in the design phase. For me, the crossover for this driver seems too high at 3.3kHz, especially with the peaks at higher frequencies. That crossover point seems more suitable for a 5" driver and Troels has a guide on driver sizes and uses. Higher order crossovers at 3.3kHz could help but it really needs the crossover point lowered which means a complete crossover redesign. I've used it between 1kHz to 2.8kHz and the best use was around 2kHz (with a DX25TG-05-4 tweeter). Crossing over higher did make the driver more edgy and harsher on the ears.
The implementation of the 0.5 woofer looks more like a hybrid between a full TMM and a 2.5 way to help increase the overall SPL. Generally they are crossed over at or below the baffle step frequency with a 1st order filter. This one looks like it needs to be dropped by an octave. This may be adding more to the higher frequency peaks even though they should be more than 18dB down. Try running without the 0.5 woofer as a trial and don't worry about the bass performance.... just listen to the mids. From the FR plot in post #59, I can't see how the 0.5 yellow is summing but I assume it must be included in the blue line which must be the sum of both woofers.
The 810921 (D2608/9130000) tweeter tends to have a rising response when used with a filter. A LCR or CR network works well to tilt it back down as well as eliminating harshness and making the L-pad more accurate. I've used it as low as 2.2kHz and combined it with 18W8531G00 and XG18WH00-08 with great success. It's a great tweeter when well implemented and it replaced a 2905-970000 on one design. To play with the tweeter I would add a CR network and adjust the L-pad to see if it makes a difference.
I don't know Troel's goals when designing this speaker and each designer has their own path so I am not criticising his design. I'm only commenting on my own personal approach with these drivers.
You may want to go over the crossover build a couple of times to make sure there are no errors. Try running each driver individually and add a driver at a time. Start from the 0.5 woofer and work up and then in reverse. That may help pin point a problem. Speaker positioning can help and try with or without toe-in. I run speakers with this tweeter without toe-in.
I still think you have chosen the wrong design for your application as this design is only suitable for medium size rooms and normal SPL's. Troels mentions driving to 95dB at 2.5m You really need a large 3-way for you application and room size.
Good luck with your journey.
I've designed and built several loudspeakers using those drivers but not both in the same design. Even so I am aware of their behaviour.
The 830875 has varied quite a bit over the years as it went from Danish to Chinese manufacture. Even the Chinese built drivers have had a few changes which could effect sonics from batch to batch, so you may be hearing a variation to the driver that was used in the design phase. For me, the crossover for this driver seems too high at 3.3kHz, especially with the peaks at higher frequencies. That crossover point seems more suitable for a 5" driver and Troels has a guide on driver sizes and uses. Higher order crossovers at 3.3kHz could help but it really needs the crossover point lowered which means a complete crossover redesign. I've used it between 1kHz to 2.8kHz and the best use was around 2kHz (with a DX25TG-05-4 tweeter). Crossing over higher did make the driver more edgy and harsher on the ears.
The implementation of the 0.5 woofer looks more like a hybrid between a full TMM and a 2.5 way to help increase the overall SPL. Generally they are crossed over at or below the baffle step frequency with a 1st order filter. This one looks like it needs to be dropped by an octave. This may be adding more to the higher frequency peaks even though they should be more than 18dB down. Try running without the 0.5 woofer as a trial and don't worry about the bass performance.... just listen to the mids. From the FR plot in post #59, I can't see how the 0.5 yellow is summing but I assume it must be included in the blue line which must be the sum of both woofers.
The 810921 (D2608/9130000) tweeter tends to have a rising response when used with a filter. A LCR or CR network works well to tilt it back down as well as eliminating harshness and making the L-pad more accurate. I've used it as low as 2.2kHz and combined it with 18W8531G00 and XG18WH00-08 with great success. It's a great tweeter when well implemented and it replaced a 2905-970000 on one design. To play with the tweeter I would add a CR network and adjust the L-pad to see if it makes a difference.
I don't know Troel's goals when designing this speaker and each designer has their own path so I am not criticising his design. I'm only commenting on my own personal approach with these drivers.
You may want to go over the crossover build a couple of times to make sure there are no errors. Try running each driver individually and add a driver at a time. Start from the 0.5 woofer and work up and then in reverse. That may help pin point a problem. Speaker positioning can help and try with or without toe-in. I run speakers with this tweeter without toe-in.
I still think you have chosen the wrong design for your application as this design is only suitable for medium size rooms and normal SPL's. Troels mentions driving to 95dB at 2.5m You really need a large 3-way for you application and room size.
Good luck with your journey.
Last edited:
I got the parts to try the Zobel suggested by System 7 as well as the LCR Pascal suggested.. The Zobel is simpler so I am going to try that first. Do I leave the L-Pad circuit in place? (This may be tough because I am basically out of room on the xover board I made.)
I also got the parts to try the 1.2 ohm and 15 ohm for the L-Pad
I also got the parts to try the 1.2 ohm and 15 ohm for the L-Pad
That was a very thoughtful reply by rabbitz. 🙂
This speaker is messed up IMO. I have frequently expressed an opinion the 6" bass plus tweeter concept is really hard.
Having done a lot of testing of the whole single 6" bass and tweeter idea, my own current opinion is that third order bass and treble filter @2.8kHz will work much better.
The 2.5 way complication is frankly unwelcome.
I think this will lead to a TMM design with a third order butterworth filter. It's quite D'Appolito really. But I'm a bit short on detail right now, because those Peerless 830875 woofers are low inductance, which makes modelling hard for me.
This speaker is messed up IMO. I have frequently expressed an opinion the 6" bass plus tweeter concept is really hard.
Having done a lot of testing of the whole single 6" bass and tweeter idea, my own current opinion is that third order bass and treble filter @2.8kHz will work much better.
The 2.5 way complication is frankly unwelcome.
I think this will lead to a TMM design with a third order butterworth filter. It's quite D'Appolito really. But I'm a bit short on detail right now, because those Peerless 830875 woofers are low inductance, which makes modelling hard for me.
Attachments
Last edited:
Leave the L-pad in place and run the CR across the tweeter +ve and - output from the crossover board. The other option is to create a small board for the L-pad and CR network.
This tweeter, just like pretty much any other 1" dome is radiating omni directionally within the presence region.
I agree... thank you Rabbitz...
I got home tonight and the parts were here for the changes that I want to make but I was too tired to de-solder etc. I did change the L_pad from 2.2 and 18 ohm to 1 ohm and 12 ohm. I immediately heard a huge difference. But they still sound bad...in fact I'm started to hate them.
What I hear that I don't like is so apparent the more I listen to them, the more apparent it is. And now it is at a lower and lower volume, to a point where I can't even listen to my favorite music at moderate levels...I mean I shut it off. (My wife took the baby to visit her brother, etc, so I have some listening time, and I don't want it.) I am thinking about bringing my ACI's back out just for some perspective.
In any case...System 7 look for a PM from me.Thanks everyone who's tried to help.
Last edited:
I thought the CR or LCR have to be in the correct spot in the signal path to work properly????
So I'm not sure were they would be placed in the signal path!
Having a few days rest without listening to anything at all will be good for giving you some perspective. Sometimes we can hear things that aren't really there or a problem if we want to. That said with the last driver change I made to my system something annoyed me right off the bat and I could not listen to anything without it ******* me off. I figured out the kinks and now it sounds great.
I seldom recommend Troels' designs to people because the majority of them have issues that don't deserve to be there. This could simply be one of those times where Troels' preferences do not match with yours.
I seldom recommend Troels' designs to people because the majority of them have issues that don't deserve to be there. This could simply be one of those times where Troels' preferences do not match with yours.
You are tormenting yourself unnecessary with all of the advices what
parts values you should use to get it sorted. Get a measurement gear
and do it yourself the proper way. Ask around what software and hardware
to use and you will be in business.
From the first time I discovered this hobby, once I saw those funny looking
graphs of impedance, FR, csd... I knew right away I gotta have these.
Make your own pizza with ingredients you like. Have faith!
Always look at yourself as a project in the future. Don't look at the very moment.
If you are riding a bike, then you know you have to look 10 m ahead to be able to
drive safely, not at your front tyre or you would soon find yourself crashed into
something.
The CR is not in the signal path but placed after the L-pad, before the tweeter and parallel to R1061. It connects across the +ve and -ve. Have a look at the CR location on the woofers, eg R2021 + C2021. system7 has a schematic early in this thread showing the CR (zobel).
OK thanks...In System 7's schematic it excludes the L-Pad....that's partially why I asked...besides the fact I understand some of this but not a lot of it...LOL
Troels website states a LCR has to be in the proper place in the signal path...I assumed that included CR...since I don't fully understand what they do!
I think measurement gear will be my next investment...I was already thinking that! Thanks for confirming!
An LCR is placed in the same location as a CR. They alter the impedance plot shape of a driver.
Last edited:
Hi Clausen,
The issue you're experiencing has evolved from being subtle to
bleeding ears. I guess the only remedy you will ever take is discard
the driver itself, I'm not suggesting it is the driver to blame, and
go with a premium product unit.
I know many would say Santana is a great artist and performer, it's
just I can't listen to his music because it hurts my brain and ears
and I have no explanation for it.
The issue you're experiencing has evolved from being subtle to
bleeding ears. I guess the only remedy you will ever take is discard
the driver itself, I'm not suggesting it is the driver to blame, and
go with a premium product unit.
I know many would say Santana is a great artist and performer, it's
just I can't listen to his music because it hurts my brain and ears
and I have no explanation for it.
Hi Clausen,
Clausen wrote in the Elsinore thread :
This means nothing, implementation is also a huge factor 🙂
You have been adviced for a lot of thing, what have you tested,
In wich order ? a summary could be interresting ...
I like the the few words from Lojzek :
"Make your own pizza with ingredients you like. Have faith!"
Clausen wrote in the Elsinore thread :
This means nothing, implementation is also a huge factor 🙂
You have been adviced for a lot of thing, what have you tested,
In wich order ? a summary could be interresting ...
I like the the few words from Lojzek :
"Make your own pizza with ingredients you like. Have faith!"
I'm just getting frustrated...in part because I have a limited amount of time, so I haven't done much yet. (Also, Pascal, that Elsinore post was after a few beers.. 🙂 ) But what bothers me about this speaker seems to have gotten worse, which I am sure is due to the familiarity my ears have developed to the issue I hear.
I went back to the original xover, and changed the L-Pad to ( 1 0hm and 12 ohm r1051 and R1061) that's all I've had time to do. I don't have room on the xover board for the other changes, so I'm going to have to figure that out!
Maybe this weekend I can try the CR and the LCR, but I'll have to figure out where I am going to put them.
(I used 12 ohm because I didn't buy any 15 ohm, which I thought I had!)
I went back to the original xover, and changed the L-Pad to ( 1 0hm and 12 ohm r1051 and R1061) that's all I've had time to do. I don't have room on the xover board for the other changes, so I'm going to have to figure that out!
Maybe this weekend I can try the CR and the LCR, but I'll have to figure out where I am going to put them.
(I used 12 ohm because I didn't buy any 15 ohm, which I thought I had!)
Also...I didn't mention this...I own my own business and I have an account which owes me several thousand dollars, and they keep shorting me on the last 3 checks, which prevents me from spending any money on this until they get caught up!!!! (That's another story, but it's adding to my overall frustration.)