Passive Monitor Thread

[Robh3606]

I started setting up to build a new system. I was lucky to be able to purchase a pair of 476Mg's and was gifted a pair of 2216Nd's by my friend 4313B. I am going to use my favorite waveguide PTH1010. It has certain advantages similar to the 2344 because it is very short and has a rapid flare compared to the other PT Waveguides and is 100x100 which works very well in my room. I started taking some preliminary measurements on a temporary baffle of the raw drivers on the PTH1010.


It actually measures very well and has a distinct advantage over the 435Be in this application. It has a rising high frequency response which makes it much easier to set-up with systems using higher sensitivity woofers. With the 435Be on this waveguide you have to trade woofer sensitivity to get a usable last octave response. Attached are plots for the 476Be no network and 435Be with nework and raw.

Rob

[Robh3606]

After the raw measurements I looked for a suitable network and chose the 9900 which uses the 476Mg. I simplified the crossover from a Charge Coupled Network and put that into LEAP. Then I copied it and added switches so I could take out the 3 notch filters to see what each did. First is the Schematic then the network response with all the notches in place.

Rob

[Robh3606]

Here is what the response changes are as you take out the notches one by one. The first is Notch 1, then Notch 2, then the Sereis Notch /HP The Blue curve shows the changes from the original circuit.

Rob

[Robh3606]

The next 2 attachements show all the changes and the last is with notches 1 and 2 removed. This became the baseline for a modified crossover.

Rob

[Robh3606]

New schematic which shows the final network for now until I can take actual baffle measurements. As you can see from the previous post the response with both notches 1 and 2 removed was not bad. What I ended up doing was changing the values of the inductor and capacitor in the series HP filter and adding a low Q 1.5 notch filter centered at 1.6K. The notch took out the bump around 2 K and the changes in the HP smoothed out the response in the upper octaves.

Attached are the results after the changes. So far they look good. I need to build up a network to see what the actual measurements look like.

Rob

[4313B]

[marconi118]

476Mg is 16ohm?

if I go with 2452HSL 8ohm, I would put 8R in series to the driver to bring it up to 16ohm and use the same crossover as yours?

[Robh3606]

476Mg is 16ohm?

if I go with 2452HSL 8ohm, I would put 8R in series to the driver to bring it up to 16ohm and use the same crossover as yours?

I don't think that's going to work. As a minimum a different driver on the M2 is going to give you a different impendence plot. Take a look at this thread where I have 8 ohm 2453-SL's on the PTH1010. There is a network and more posted. Might be a better place to start as at least these are 8 ohm drivers and you can compare the Imp plots in this thread with the 476Mg to see how different they can be on the same waveguide.

That thread just does the CD comp and a protection cap for use in an active crossover system. You could drop a high pass in front but you would have to tweak it a bit for it to work.

http://www.audioheritage.org/vbullet...ng-Impressions

Rob

[Ian Mackenzie]

476Mg is 16ohm?

if I go with 2452HSL 8ohm, I would put 8R in series to the driver to bring it up to 16ohm and use the same crossover as yours?

In typically the drivers labeled 8 ohms have an impedance of 6-8 ohms and those labeled 16 ohms have an impedance of 12 ohms. You could fiddle around with a simple passive EQ and get an okay result. Care needs to be taken with passive EQ that it does not load down the impedance too much at high frequencies using a capacitor shunt across a series resistance. A small value inductor (0.3-0.5 mH) with a small value series power resister (2-3.9 ohms) shunted in parallel with the driver would be a good starting point. Greg Timbers has used this approach and it acts to protect the driver. This approach does not load down the impedance. It attenuates frequencies below the shunt resonance of inductor at 6 dB per octave. Then you need to have a high and low pass crossover filter.

But beyond that it’s a lot of trial and error with test microphone. Dayton make a loudspeaker measurement set up. An active crossover with your diy passive EQ is a good option.

Alternatively you might want to try the Dayton dsp crossover first to get yourself up and going as a Diy effort. Tip. When you actively EQ a driver cut the frequencies below 3-6 khertz in preference to boosting the high frequencies. This will improve the noise and distortion performance of the Dsp processor.