EDINGDALE speaker build

Dedicated to those large boxes at one end of the room
Post Reply
User avatar
Toppsy
Shed dweller
Posts: 2400
Joined: Mon Jul 23, 2007 12:10 pm
Location: red rose country

#76

Post by Toppsy »

OK. I managed to find some time last night and the night before to have a go taming the sibilance we all heard my place last Sunday.

Wednesday night I tried adding a small inline 0.68Ω 5W resistor to the MA tweeter. This did take some of the edge off the sibilance but not enough on the most offending tracks. So I replaced this with a 1Ω 10W jobbie. This did cure the sibilance BUT took some of the life out the top end sparkle we had before.

Chris came round later that night to drop off the L-pad attenuators he had borrowed for his FH3 / CHP70 / T#3 experiments and agreed with my observations. As nice a sound as it was it just lacked that certain edge and presence. As nice as the little MA tweeter is I do think it benefits greatly if not driven at full SPL as we have it here. Attenuating the tweeter with the inline resistor proved this but because of this SPL limit of the tweeter with the attenuation fitted it was being swamped a little by the slightly greater efficiency of the combined woofers. If you have a mid-woofer of 90dB or less efficiency then padding down the MA tweeter to this will work a treat.

So armed with this partially successful experiment, last night I rigged up a temporary housing for the Monocor ribbons using whatever scraps off-cuts of MDF I had to hand. This was wired through the aforementioned L-Pad attenuators. See below photos.

Now this is where it became rather interesting. With around 5dB dialled on the L-Pads the top end sparkle was back in buckets full of top end presence. No sibilance to talk about, just the slightest hint on one track, but I think that is in the recoding itself and not a fault of the ribbons. These are smooth as silk and the whole soundstage upped a notch or too. I'm well impressed. Chris was present for a second opinion and had to agree that in this configuration of driver arrangement and in these cabinets the ribbons are a big step up in performance. We did play around with different levels of attenuation but in the end both agreed somewhere between -5dB and -6dB was just about right in my room.

So now I have to modify the front baffles to replace the MA tweeters with the Monocor Ribbons, mount the L-Pad attenuators on the back baffle and rewire back to the XO.

This I hope to find the time to do over this coming weekend. However, before I can do this I have a FH3 flat-pack kit to put together ready to ship out to Norway next week. Busy, busy busy, but I'm not complaining.

Oh! And thanks to NickG, or was it Ali, who suggested we try a ribbon on these speakers. Good call mate. :lol: :lol:

Image Image
Last edited by Toppsy on Fri Sep 30, 2011 6:56 pm, edited 1 time in total.
User avatar
planet10
Old Hand
Posts: 484
Joined: Sun Sep 23, 2007 4:32 pm
Location: The Colonies
Contact:

#77

Post by planet10 »

Colin,

The sibilance is probably from XOing the tweeter too low without enuff filter.

dave
User avatar
Toppsy
Shed dweller
Posts: 2400
Joined: Mon Jul 23, 2007 12:10 pm
Location: red rose country

#78

Post by Toppsy »

Dave,
I take your point.

However, the XO to the MA tweeter is 3.5K and is a second order filter. This works a treat on the Edale 2-way standmounts. This has a efficiency of around 87dB ie that of the mid-woofer that peaks at around 5K from the FR plots. With the 4 mid-woofers the efficiency is upped to around 93dB. Crossing over higher I think would move the FR of the woofers into an area that may not be best for them.

The MA T#3 (ERT26) maxs out at around a published 92dB. And it is here I think the tweeter is struggling. With added load like the adding of the resistor this tames the sibilence but in our case also reduces the overall efficiency by about 1 dB and I think this is the issue for the T#3 in this instance. As I say with the T#3 padded down to a mid-woofer of 90dB efficency or lower, using the same XO filter circuit the tweeter sounds wonderful. So I can only assume it's down to trying to ask a little too much from this fine (and cheap) unit.

I could be wrong. I let me ears do the talking.

EDIT.

Having just checked out the FR plots for the mid-woofers it would appear I could push the XO point to a max 4.5kHz. I'll have to see what the XO componets changes would need to be for this and see what suitable components I have in spares box to try this out.
Last edited by Toppsy on Fri Sep 30, 2011 9:50 am, edited 1 time in total.
User avatar
planet10
Old Hand
Posts: 484
Joined: Sun Sep 23, 2007 4:32 pm
Location: The Colonies
Contact:

#79

Post by planet10 »

Adding the series R both reduces its level and pushes the XO frequency up.

dave
User avatar
Toppsy
Shed dweller
Posts: 2400
Joined: Mon Jul 23, 2007 12:10 pm
Location: red rose country

#80

Post by Toppsy »

Adding the series R both reduces its level and pushes the XO frequency up.
Perhaps. BUT for the small cost increase in the price I pay for the ribbons the improvement in overall sound far outweighs the small increase in costs.

Given the choice of paying £10 more for the ribbons over the silk domes, which would you choose if offered the option?

And rightly or wrongly to your average punter a Ribbon Tweeter is far more exotic and appealing alternative over a simple (though very good, IMO) silk done.

However, if I have the components in my spares box I shall try a revised XO say at 4K or 4.5K depending what I have to hand.
Last edited by Toppsy on Fri Sep 30, 2011 11:52 am, edited 1 time in total.
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#81

Post by Nick »

planet10 wrote:Adding the series R both reduces its level and pushes the XO frequency up.

dave
Does it? I would have thought increasing the load resistance would have lowered the crossover frequency.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
chris661
Shed dweller
Posts: 2556
Joined: Mon Feb 22, 2010 7:29 am
Location: Sheffield

#82

Post by chris661 »

Nick wrote:
planet10 wrote:Adding the series R both reduces its level and pushes the XO frequency up.

dave
Does it? I would have thought increasing the load resistance would have lowered the crossover frequency.
Try it yourself - it's more complicated...

http://www.the12volt.com/caraudio/crosscalc.asp
User avatar
Toppsy
Shed dweller
Posts: 2400
Joined: Mon Jul 23, 2007 12:10 pm
Location: red rose country

#83

Post by Toppsy »

Try it yourself - it's more complicated...
I take it by that you're not sure yourself otherwise you would have given actual results of adding the series inline 1 ohm resister. :wink:

I'd like to see a working example of the maths; That adding the small resistor reduces the level (by approx 1 dB in this case) and pushes up the XO frequency.

However, all this is rather arbitrary, as after another listen late this afternoon comparing the two tweeters, the ribbon wins hands down in this design.

So having now 90% completed the construction of the FH3 flat pack kit destined for Norway, I shall this weekend attack the front baffles of the Edingdales to install the ribbons in place the MA tweeters. Simples. :lol:
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#84

Post by Nick »

chris661 wrote:
Nick wrote:
planet10 wrote:Adding the series R both reduces its level and pushes the XO frequency up.

dave
Does it? I would have thought increasing the load resistance would have lowered the crossover frequency.
Try it yourself - it's more complicated...

http://www.the12volt.com/caraudio/crosscalc.asp
Well, I can't see what that page is trying to tell me. Wiki suggests that the corner frequency of a high pass LCR filter is not related to R. Do you have any equations instead of a web page?

It seems counterintuitive. If it was just a CR filter, increasing the load resistance would certainly lower the corner of the high pass filter.

f = 1/2*pi*R*C

increasing R will lower f.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Ali Tait
Eternally single
Posts: 4373
Joined: Fri Jun 08, 2007 8:10 pm
Location: Galashiels

#85

Post by Ali Tait »

Good news Colin, look forward to hearing the speakers with the ribbons.
chris661
Shed dweller
Posts: 2556
Joined: Mon Feb 22, 2010 7:29 am
Location: Sheffield

#86

Post by chris661 »

Nick wrote:
chris661 wrote:
Nick wrote:
Does it? I would have thought increasing the load resistance would have lowered the crossover frequency.
Try it yourself - it's more complicated...

http://www.the12volt.com/caraudio/crosscalc.asp
Well, I can't see what that page is trying to tell me. Wiki suggests that the corner frequency of a high pass LCR filter is not related to R. Do you have any equations instead of a web page?

It seems counterintuitive. If it was just a CR filter, increasing the load resistance would certainly lower the corner of the high pass filter.

f = 1/2*pi*R*C

increasing R will lower f.
Scroll down a little, there's some equations.


L1 (mHy) = (1000 x Impedance) / (6.283 x Crossover Frequency)
C1 (µfd) = 1,000,000 / (6.283 x Impedance x Crossover Frequency)

The above are for first order crossovers, (for 2nd order, you multiply the results by 1.414 and 0.707 respectively).
As you can see, increasing the impedance for the capacitor will decrease the crossover frequency.
For inductors, it's the opposite.

I'm fairly sure Colin has a 2nd order filter. So the actual crossover points (for cap and inductor) will diverge, giving an unintended result.

Methinks to avoid such messing around, a true L-pad should be used: one resistor in parallel with the speaker, another in series to keep the impedance the same as the original.

Chris
User avatar
Nick
Site Admin
Posts: 15707
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#87

Post by Nick »

OK, I meant real LCR equations, but they will do
C1 (µfd) = 1,000,000 / (6.283 x Impedance x Crossover Frequency)
Ok, rearange

Cross over frequency = 1,000,000 / ( 6.283 * C1 * Impedance)

So as I said. INCREASING the impedance will LOWER the crossover frequency.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
chris661
Shed dweller
Posts: 2556
Joined: Mon Feb 22, 2010 7:29 am
Location: Sheffield

#88

Post by chris661 »

Nick wrote:OK, I meant real LCR equations, but they will do
C1 (µfd) = 1,000,000 / (6.283 x Impedance x Crossover Frequency)
Ok, rearange

Cross over frequency = 1,000,000 / ( 6.283 * C1 * Impedance)

So as I said. INCREASING the impedance will LOWER the crossover frequency.
For capacitors, you are absolutely right. I never said otherwise.

Rearrange the equation for the inductors.

Frequency=impedance/2pi*inductance
Frequency is proportional to impedance.

Being a 2nd order crossover, shouldn't both be considered?
User avatar
pre65
Amstrad Tower of Power
Posts: 21373
Joined: Wed Aug 22, 2007 11:13 pm
Location: North Essex/Suffolk border.

#89

Post by pre65 »

I thought the "discussion" was about the effect of resistors ? :?
The only thing necessary for the triumph of evil is for good men to do nothing.

Edmund Burke

G-Popz THE easy listening connoisseur. (Philip)
User avatar
planet10
Old Hand
Posts: 484
Joined: Sun Sep 23, 2007 4:32 pm
Location: The Colonies
Contact:

#90

Post by planet10 »

Toppsy wrote: BUT for the small cost increase in the price I pay for the ribbons the improvement in overall sound far outweighs the small increase in costs.
I have no disagreement there. What i supposed to be implicit in what i was saying was that the MA wasn't suited to this application

dave
Post Reply