Hi All,
I have recently acquired a pair of Arcaydis EB2s speakers.. Slightly unconventional in their configuration, with a 20l sealed box design.
EB2S – Arcaydis
I was thinking of upgrading the mid/woofer to a higher-spec unit, but I am still learning about driver parameters and their effect on the sound.
The woofer used is the visaton W 170 S --- crossed over at 2.3kHz
187mm / 148mm
W 170 S - 8 Ohm
This has a pretty flat response, high .QTS and a 86db/1w sensitivity... I was hoping to find a unit with similar specs to avoid having to re-do the crossover, but finding a unit with high QTS for use in a sealed box ain't easy!
The closest I have come so far is this driver; The WaveCor WF182DB10
http://manuals.audiomania.ru/data/wavecor_wf182bd09-01.pdf
Seems to behave similarly and has the right sensitivity - but QTS is down at 0.37
Can anyone suggest the effect this will have on the sound? Right now the speaker is a trifle fierce across the mids, not harsh but without much warmth
Thanks for any tips
I have recently acquired a pair of Arcaydis EB2s speakers.. Slightly unconventional in their configuration, with a 20l sealed box design.
EB2S – Arcaydis
I was thinking of upgrading the mid/woofer to a higher-spec unit, but I am still learning about driver parameters and their effect on the sound.
The woofer used is the visaton W 170 S --- crossed over at 2.3kHz
187mm / 148mm
W 170 S - 8 Ohm
This has a pretty flat response, high .QTS and a 86db/1w sensitivity... I was hoping to find a unit with similar specs to avoid having to re-do the crossover, but finding a unit with high QTS for use in a sealed box ain't easy!
The closest I have come so far is this driver; The WaveCor WF182DB10
http://manuals.audiomania.ru/data/wavecor_wf182bd09-01.pdf
Seems to behave similarly and has the right sensitivity - but QTS is down at 0.37
Can anyone suggest the effect this will have on the sound? Right now the speaker is a trifle fierce across the mids, not harsh but without much warmth
Thanks for any tips
Last edited:
Hi,
Less bass in terms of absolute depth extension or a general more anaemic sound across the board?
The res. freq of the wavecor is lower - could this help 'compensate' for the qts disparity?
Less bass in terms of absolute depth extension or a general more anaemic sound across the board?
The res. freq of the wavecor is lower - could this help 'compensate' for the qts disparity?
I can only think of the SS classic 18w, around the same sensitivity. Expensive but they come up on ebay and in classifieds quite often.
20L of box with a 6" reflex or a closed box driver is actually about right.
Which works best, depends on the driver Qts. The W 170 S - 8 Ohm comes in 4 and 8 ohm versions, and you should measure yours to see what you have.
Replacing it would be my last choice. There's nothing wrong with it:
Here's a project we did with similar drivers:
http://www.diyaudio.com/forums/multi-way/305201-visaton-2-tower-monitor-audio-bronze-5-a-11.html#post5070435
That is just one of about a dozen ways to do a filter for this style. We went for a flat impedance idea just for the interest. But the filter works at every level IMO:
What I would do, in your position, is evaluate the crossover thoroughly. Draw the schematic. Then we can see if there are any glaring faults with it. Or how specifically tailored it is to the driver. Easy enough to sim your speaker: Downloads.
Changing the woofer might be the least interesting thing you can do. You might, for instance, find that a different style of tweeter does far more to make the sound more to your taste. And notching or reducing the 5kHz cone breakup is very audible.
Which works best, depends on the driver Qts. The W 170 S - 8 Ohm comes in 4 and 8 ohm versions, and you should measure yours to see what you have.
Replacing it would be my last choice. There's nothing wrong with it:
Here's a project we did with similar drivers:
http://www.diyaudio.com/forums/multi-way/305201-visaton-2-tower-monitor-audio-bronze-5-a-11.html#post5070435
That is just one of about a dozen ways to do a filter for this style. We went for a flat impedance idea just for the interest. But the filter works at every level IMO:
What I would do, in your position, is evaluate the crossover thoroughly. Draw the schematic. Then we can see if there are any glaring faults with it. Or how specifically tailored it is to the driver. Easy enough to sim your speaker: Downloads.
Changing the woofer might be the least interesting thing you can do. You might, for instance, find that a different style of tweeter does far more to make the sound more to your taste. And notching or reducing the 5kHz cone breakup is very audible.
Not a bad suggestion if you think that replacing one paper 6" woofer with another will make a huge difference:
H1217-08 CA18RLY
W 170 S - 8 Ohm
IMO, these theories are a bit lazy. I presume the original poster is a bit dissatisfied with his speaker. IMO, it's a question of rolling your sleaves up and investigating what you have bought here.
Anyone who thinks there is such a thing as a perfect speaker really ought to go back to school. The best we can do is something that might fool you. Now show me some effort. It's hard work. It's compromise. Our worst enemy is the room. It's not the same as the World. 😀
Thanks for the suggestions.
The SEAS driver is up 4db on the visaton at the crossover frequency - 2.3KHz - so I didn't think that'd work, unless I changed a resistor on the tweeter or something.
The Scan Speak 18W/8531G00 looks like a contender but a very spendy one.
The Eton 8-412 looks like another option
Eton 8-412/C8/32 HEX Symphony II 8" woofer
Albeit one that'd require some cabinet modifications to fit.
The SEAS driver is up 4db on the visaton at the crossover frequency - 2.3KHz - so I didn't think that'd work, unless I changed a resistor on the tweeter or something.
The Scan Speak 18W/8531G00 looks like a contender but a very spendy one.
The Eton 8-412 looks like another option
Eton 8-412/C8/32 HEX Symphony II 8" woofer
Albeit one that'd require some cabinet modifications to fit.
D Marsden, crossovers are specific to drivers when well executed. If you draw up the schematic, or post a photo of it with the paperclip thingie, you can then have an idea what can be fixed.
That W170S is a very good woofer by all accounts. It often benefits from a notch around 5.5kHz, as do most 6" drivers. That is simple enough to do. Needs about a 15R resistor and 1uF capacitor across the bass coil. Parts cost about £5 at Maplin.
And, FWIW, once I see the filter, I can have a simulation running in about 5 minutes, which might reveal what your ears aren't liking. Commercial product often skimp on filters.
That W170S is a very good woofer by all accounts. It often benefits from a notch around 5.5kHz, as do most 6" drivers. That is simple enough to do. Needs about a 15R resistor and 1uF capacitor across the bass coil. Parts cost about £5 at Maplin.
And, FWIW, once I see the filter, I can have a simulation running in about 5 minutes, which might reveal what your ears aren't liking. Commercial product often skimp on filters.
Last edited:
That looks pretty reasonable, except for dreadful coil layout. Might be a BW3 design:
Ignore the red impedance correction.
You need to trace the circuit, and get the capacitor and 4 band resistance values from the standard colour codes. I can probably wing it on the coils.
Ignore the red impedance correction.
You need to trace the circuit, and get the capacitor and 4 band resistance values from the standard colour codes. I can probably wing it on the coils.
Attachments
Last edited:
Steve, could I get your expert advice to reach a conclusion on this old thread? It might help other EB2S owners like me.That looks pretty reasonable, except for dreadful coil layout. Might be a BW3 design:
Ignore the red impedance correction.
You need to trace the circuit, and get the capacitor and 4 band resistance values from the standard colour codes. I can probably wing it on the coils.
I shared Paul Messenger’s assessment of them:
‘listening to music programming at higher volume levels, did have a degree of ‘forwardness’ that could become a tad wearing, depending upon the mix.’
The majority of the time I’m not listening at higher volume levels so it’s only a minor irritation. It could be a down to the room acoustics but Paul Messenger’s review did make me wonder. Based on his in-room far-field averaged measurements he noted:
‘A very slight trend in the overall output maximises output above 600Hz while minimising output below that point, which possibly helps project vocals slightly.’
Other than that he was impressed by his measurements:
‘Much of the band holds comfortably within ±3dB (or even tighter!) limits, so superior neutrality is guaranteed. The bass extension is pretty good too, registering -2dB at 30Hz and -9dB at 20Hz under our far-field measure, alongside a relatively even, albeit slightly dry balance. For the record, the main driver/box resonance occurs here at 64Hz.’
He also measured the impedance as ‘remaining above eight ohms throughout’. Though EB2S specs give minimum impedance as 5.9ohms. That is the value for Rdc in Visaton’s specs for the 8ohm W 170 S. The tweeter is Monacor DT-99 btw.
I’m not looking to replace either of the drivers. I would consider modifying/changing the crossover if there were glaring faults. Your view in #11 was that it looked reasonable apart from dreadful coil positioning. So that might point to some judicious use of EQ software to make a fix.
From the photo taken by original poster it’s difficult to make out the values on the three caps. R1 & R2 both appear to be 6.8 ohms (if they are blue/grey/gold/gold). It looks like (to a rank amateur) that the crossover components for W 170 S is one coil and one cap (L3 & C4); and for tweeter there are 2 coils (L1 & L2), 2 caps (C1 & C2) and 2 resistors (R1 & R2). Tracing the circuit as you requested would involve unscrewing the crossover which, as a complete DIY beginner, I’m wary of doing.
To demonstrate some rolling up of sleeves I did look at the Claritycap website. Their PWA range are 160Vdc. Going by values available there (http://www.claritycap.co.uk/products/pwa.php) C2 could be 22uF and C4 could be 2.2uF. Printing on C1 is obscured by red and black wires. Puzzled by lack of C3?
Are you able wing it completely with this (albeit sketchy) additional info?
Philip
You missed the point. Dreadful here means that there could be some cross talk between the coils, not that there is something inherently faulty in the crossover that needs to be corrected.
Frankly, a review without shown measurements is pretty useless. If you want to modify the crossover you need first to trace it. We - not EB2s owners - don't know the details, we can only make assumptions. And then a FR would be pretty essential, as without it making mods involves a good hearing in knowing what is wrong and where is located (frequency wise).
Ralf
Unscrewing a crossover board is, as the word implies, simply unscrewing it from the cabinet
These are probably fine speakers/drivers (as is the driver W170S) and it won´t be easy to make a new crossover without experience, measurement equipment and the willingness to invest in some inductors and caps. Even more difficult to end up with something better than the original.
But if you like the idea of tackling that yourself I´m sure you´ll get some help here.
If I was in your situation I´d measure the speakers first and see if they can be improved.
I´d bet the XO and/or tweeter is the weakest link but I can just as well be wrong. Nothing wrong with a sealed W170S though.
(oh and can you rule out that it is your room maybe?)
I guess there´s lots of designs out there with W170S+G20SC/SC10N that are pretty good.
System7 alone must have posted a plethora of Visaton designs here that could fill a book on their own; plus many found in the Visaton forums.
With regards to woofers, you can use speakerboxlite which is an online simulation tool.
You can click "choice", then define "sealed" or "vented" and give a box volume.
The database is full with old drivers too and also car drivers so you´ll have to scroll a bit but will find many known drivers amongst the results:
Speakerboxlite
If you fancy that Wavecor and rather have a finished XO-design these might fit the bill:
DXT-MON 182
Not sure though the XO will be cheap plus the drivers aren´t cheap plus the plans cost 60€ or so.
The guy might use lots of parts to arrive at that ruler flat response
(which BTW is not to everyones liking). These are highly regarded though.
But even if you go for that I´d keep/sell your speakers in its original state and build something from scratch.
The monitors will benefit from that front baffle shape too.
But if you like the idea of tackling that yourself I´m sure you´ll get some help here.
If I was in your situation I´d measure the speakers first and see if they can be improved.
I´d bet the XO and/or tweeter is the weakest link but I can just as well be wrong. Nothing wrong with a sealed W170S though.
(oh and can you rule out that it is your room maybe?)
I guess there´s lots of designs out there with W170S+G20SC/SC10N that are pretty good.
System7 alone must have posted a plethora of Visaton designs here that could fill a book on their own; plus many found in the Visaton forums.
With regards to woofers, you can use speakerboxlite which is an online simulation tool.
You can click "choice", then define "sealed" or "vented" and give a box volume.
The database is full with old drivers too and also car drivers so you´ll have to scroll a bit but will find many known drivers amongst the results:
Speakerboxlite
If you fancy that Wavecor and rather have a finished XO-design these might fit the bill:
DXT-MON 182
Not sure though the XO will be cheap plus the drivers aren´t cheap plus the plans cost 60€ or so.
The guy might use lots of parts to arrive at that ruler flat response
(which BTW is not to everyones liking). These are highly regarded though.
But even if you go for that I´d keep/sell your speakers in its original state and build something from scratch.
The monitors will benefit from that front baffle shape too.
Last edited:
re:'hoping to find a unit with similar specs to avoid having to re-do the crossover" - the crossover has nothing to do with the specs, it's based on the impedance plot
Quite happy with my Seas ER18's but the SBs I just used are quite nice. Purify and CSS worth looking at. I find ScanSpeak to be good, but often hard to use crossover wise.
Grab a copy of WinISD and simulate to your hearts content.
Grab a copy of WinISD and simulate to your hearts content.
Replace a driver and the individual drivers will need to be measured for a new crossover.
If you don't like them, you're far better off selling them.
If you don't like them, you're far better off selling them.
I just had a look at this, based on the complete lack of crossover tracing that has been done so far. Seem to pick out two 6.8R resistors, a 8.2uF on the bass and possibly 4.7uF and 12uF on the tweeter.
15 minutes work in Boxsim 1.2 and I might have found the problem of forwardness here. But a compete guess on the filter. So I really can't guarantee anything.
If that is a third order bass filter, the 8.2uF shunt needs some resistance added, I tried 6.8R. Otherwise it's very peaky at 1.8kHz.
It's a 92dB 100mm tweeter as it goes: https://www.monacor.com/products/components/speaker-technology/hi-fi-tweeters-/dt-99/
Familiar enough: https://www.diyaudio.com/community/...sealed-simple-loudspeaker-build-plans.352046/
TRACE THE CIRCUIT! OTHERWISE WE ARE GUESSING....
15 minutes work in Boxsim 1.2 and I might have found the problem of forwardness here. But a compete guess on the filter. So I really can't guarantee anything.
If that is a third order bass filter, the 8.2uF shunt needs some resistance added, I tried 6.8R. Otherwise it's very peaky at 1.8kHz.
It's a 92dB 100mm tweeter as it goes: https://www.monacor.com/products/components/speaker-technology/hi-fi-tweeters-/dt-99/
Familiar enough: https://www.diyaudio.com/community/...sealed-simple-loudspeaker-build-plans.352046/
TRACE THE CIRCUIT! OTHERWISE WE ARE GUESSING....
Attachments
I appreciate you taking time to run the FR prediction in Boxsim 1.2 and for your educated guess on crossover layout and component values.
I was able to get the crossover circuit diagram from one of the partners in Arcaydis Audio (which is no longer trading but he is still in the business – Matrix Pro-audio).
Look how close you were with your guesswork!
All the values on tweeter filter are correct – only difference is R2 is in series with C2. It is a third order bass filter and your C4 value is correct – just slightly out on the coil values: L2 is 0.45mH and L3 is 2.2mH.
Now we are in the know can we conclude that the problem of forwardness is the peaky response at 1.8kHz as you suspected? Are the different coil values on the bass filter (or different position of R2 on tweeter filter) likely to alter the FR predicted by Boxsim? If so would you mind running it again please as my initial plan is to address the problem with EQ software (FFMPEG equaliser) and I’d need to have an idea of the height and width of predicted peak at 1.8kHz, and if any other frequencies need correcting.
If didn’t have option to use EQ software (e.g. I used a CD player) then I’d be looking to add a resistor to the bass shunt as you recommended. A couple of questions about doing that: (a) would you still say 6.8R? and (b) how easy would it be to fit that resistor given that space on pcb is tight?
One other question about Boxsim. Can the FR prediction using the SC 10N as the tweeter in Vistaton range which is nearest to Monacor DT-99 be taken as reliable indicator of what it will be for EB2S speakers? Does the higher sensitivity of the DT-99, or any other differences in its parameters, mean that actual FR will differ significantly from Boxsim prediction?