You can try to measure the can size and look at SCr/Solen catalog and Clarity Caps (PX or ESA?) if you have a match.
How much time has this loudspeaker, 10 years ? The Elcap measure exactly the marked value ?
I ask because sometimes they sort out the capacitance or a drift can occurs with aging if more than a decade is possible (which is perhaps not a problem at this 18 uF location, btw maybe wanted not too much low esr there)
If the mkp caps are exactly the marked values then it is likely more to be claryties or any brand in the 2 to 3% precision like Audyn. Though what you are describing is not too much an Audyn cap. You may have an idea by measuring the caps case.
Just guess of course and if your sparkles that are missing are not comming from the source or an atribute of the XO.
The first tips given are cheap to try : cap // with the 1.5 resistor, or reduce a little the resistor value, or swapping with a resistor that is a little less sweet that the cement wirewound. I would try the first two experiment as the cheapest for understanding the XO behavior and what can be made later before you spent monney on tweeter that is a little a shot in the dark.
My two cents only.
How much time has this loudspeaker, 10 years ? The Elcap measure exactly the marked value ?
I ask because sometimes they sort out the capacitance or a drift can occurs with aging if more than a decade is possible (which is perhaps not a problem at this 18 uF location, btw maybe wanted not too much low esr there)
If the mkp caps are exactly the marked values then it is likely more to be claryties or any brand in the 2 to 3% precision like Audyn. Though what you are describing is not too much an Audyn cap. You may have an idea by measuring the caps case.
Just guess of course and if your sparkles that are missing are not comming from the source or an atribute of the XO.
The first tips given are cheap to try : cap // with the 1.5 resistor, or reduce a little the resistor value, or swapping with a resistor that is a little less sweet that the cement wirewound. I would try the first two experiment as the cheapest for understanding the XO behavior and what can be made later before you spent monney on tweeter that is a little a shot in the dark.
My two cents only.
Last edited:
A few of us have already suggested bypassing the resistor. If you use a fairly big cap, say 33uF, the filter slope remains the same, while the 10kHz region gets lifted 1,5dB. That should just bring the sparkle. See #174 😉
Although I would have a hard time justifying it, given tone?'s interest in a tweeter upgrade to the Scan D3004/6640, here is how it might look compared to the current one. Take this with a large grain of salt though tone? as there is no guarantee that the files I am using are correct. For the Scan D2604, I'm using Zaph's measurements and for the D3004, I'm using HiFi Compass's. Files do not contain baffle diffraction responses but it shouldn't matter as they should be exactly the same and so therefore cancel each other out for the purposes of comparison.
Red is the current tweeter with the correct xo values. Blue is the D3004 with xo adjustments and black is the same with the addition of another capacitor to increase the top end 'sparkle' zone.
And the filters:
So it looks doable to me.
But for better accuracy, you should take measurements of the raw (no xo) FR and impedance of the current tweeter on the baffle and then repeat the exact same thing with any new tweeter, again on the baffle.
Red is the current tweeter with the correct xo values. Blue is the D3004 with xo adjustments and black is the same with the addition of another capacitor to increase the top end 'sparkle' zone.
And the filters:
So it looks doable to me.
But for better accuracy, you should take measurements of the raw (no xo) FR and impedance of the current tweeter on the baffle and then repeat the exact same thing with any new tweeter, again on the baffle.
i am assuming that i would need special software to measure the tweeter on the baffle ?
i know Dayton audio has that software, but I have a mac. Ugh
also I know you are gonna be ticked, but someone mentioned that maybe the D3004/6600 tweeter would be a better match for my drivers and less of an ‘ investment’.
pardon, I really appreciate the work you have done above.
Right now since I found the values of the xo, I need to build an external xo, with better parts, and also to use if I change the tweeter.
i need to make a shopping list for the external xo.
i need to make a shopping list for the external xo.
Nice post, jreave, I feel the better way to be accurate without redesigning is closer to your first way. It's simpler too, not needing measurements. As you say, the baffle effects remain, it's just the differences between the tweeters that counts. This assumes tone? wants to make them comparable to start with. It should only take factory response and impedance plots and a simulator.
yeah comparable sounds good.
im just expecting the D3004/6600 to have a much nicer sound than my current tweeter. many manufactures use that tweeter and swear by it.
i don’t think I would want it crossed-over to a different freq rather than 2.5khz where it is crossed over now. I think that might change the character of the speaker too much. I could experiment with that later on of course.
im quoting the 2.5khz btw from measurements I took of my speakers. Not the above chart
don’t know if that makes sense given my very limited knowledge
ill try and find them.
I’m assuming you want these?
https://www.scan-speak.dk/datasheet/pdf/d3004-660000.pdf
https://www.scan-speak.dk/datasheet/pdf/d2604-830000.pdf
below is a measurement I took of my speaker at one foot with the mic at the center of the tweeter if it helps. I’m assuming that dip at around 4khz is the baffle Diffraction?
I’m assuming this software would be ideal for me to do this perfectly?
https://daytonaudio.com/product/1650/dats-v3-computer-based-audio-component-test-system
https://daytonaudio.com/product/1650/dats-v3-computer-based-audio-component-test-system
And fellas , please be patient with me.
i am just a novice, so I taking shots in the dark right now.
thanks for understanding.
i am just a novice, so I taking shots in the dark right now.
thanks for understanding.
im Assuming these aren’t for me since I don’t have the software?
and I assuming you made these yourself with the info I provided? In which case I thank you
You can use them in any crossover simulator. Here I've shown how it would look if you did nothing to the crossover, just screw in the other tweeters. The sim uses the respective impedance plots. The current d2604 is in red.
It appears the first thing to try after this is reducing the resistance...
It appears the first thing to try after this is reducing the resistance...
yeah someone sent me a pad calculator last night and doing the math I think changing the resistor to 1 ohm ( instead of the 1.5ohm currently) for the 3004/6600 tweeter would work nicely.