Crossover component selection

ketchup · « **on:** 6 Jan 2023, 04:21 pm »

There seems to be a good amount of crossover knowledge here, so I hope it's alright to ask for advice about parts selection for rebuilding my internal Magnepan 3.6R mid/high crossovers with better parts. I biamp, so I don't use the external crossover. All I'll be rebuilding is this. I'd like to keep the cost, excluding wire and connectors, to under $1000. If that's not possible or if there is significant improvement to be had by spending more, I can stretch the budget.

Since I don't know much about what caps and inductors sound good, I could use some help. I'd like to lean away from bright sounding parts since the system probably tilts slightly in that direction already. Parts that lean slightly toward lush or rich sounding would be preferred as long as they don't sacrifice necessary detail. Any size components are okay as I can mount them externally if they don't fit within the original crossover cavity of the panels. Any help would be greatly appreciated.

Badd99 · « **Reply #1 on:** 6 Jan 2023, 05:31 pm »

Badd99 · « **Reply #2 on:** 6 Jan 2023, 05:32 pm »

I would replace that stack on the tweeter to a 18uF vcap odam. This is a significant improvement and worth every cent. Bypass if you like a warmer sound the 0.01 pure copper duelund cap.

For the mids I'd use 25uF sonicap gen 1s. I personally would bypass these with 0.1 miflex pure copper caps. Made a difference even though a shunting cap for my nx otica.

Resistors I'd use path audios resistors. Some had magnetic problems but I've bought 9 of them and none have an issue at all with neo magnets. Way better sounding than the other "Premium" resistors in my test.

I would wire with solid core wire, Danny's wire on his website is a good deal and good wire.

Inductors I would foil inductors on the mids as they are in line of the signal and air core on the tweeter if you wanted to save money. Personally I'd do all copper foil. Dayton or Janzen inductors are good for this project.

Hafgrim · « **Reply #3 on:** 6 Jan 2023, 06:13 pm »

https://www.humblehomemadehifi.com/Cap.html

Here is this guide I like to keep an eye on.

ketchup · « **Reply #4 on:** 6 Jan 2023, 08:29 pm »

Quote from: Badd99 on 6 Jan 2023, 05:32 pm

I would replace that stack on the tweeter to a 18uF vcap odam. This is a significant improvement and worth every cent. Bypass if you like a warmer sound the 0.01 pure copper duelund cap.

For the mids I'd use 25uF sonicap gen 1s. I personally would bypass these with 0.1 miflex pure copper caps. Made a difference even though a shunting cap for my nx otica.

Resistors I'd use path audios resistors. Some had magnetic problems but I've bought 9 of them and none have an issue at all with neo magnets. Way better sounding than the other "Premium" resistors in my test.

I would wire with solid core wire, Danny's wire on his website is a good deal and good wire.

Inductors I would foil inductors on the mids as they are in line of the signal and air core on the tweeter if you wanted to save money. Personally I'd do all copper foil. Dayton or Janzen inductors are good for this project.

Thank you for this. What effect would putting an 18uF cap at the tweeter instead of the factory-specified 17.1uF have?

Danny Richie · « **Reply #5 on:** 6 Jan 2023, 09:14 pm »

This is typical of what I have seen with many other Magnepan models that I have worked on.

The drivers are all playing on top of each other. The end result is not good.

Check out the video that I did on the 3.7i and it will illustrate and explain what I am talking about: https://youtu.be/_bpDP0jxj4k

I would solve those problems before spending money on better crossover parts.

ketchup · « **Reply #6 on:** 6 Jan 2023, 09:56 pm »

Thanks, Danny. I watched that video the other day, but unfortunately I don't have the resources or desire to completely re-engineer the crossovers. The speakers sound phenomenal as they are. I just want to squeeze a little more out of them by replacing these crossover parts and eliminating all of the steel connectors.

ketchup · « **Reply #7 on:** 7 Jan 2023, 07:19 pm »

It looks like my speakers only have 17uF at the tweeter (there is no 0.1uF cap). In theory, how will 18uF in this location effect the crossover? Are there any other premium caps that will work well in this location that are 17uF?

Hobbsmeerkat · « **Reply #8 on:** 7 Jan 2023, 08:38 pm »

18uF is about 6% larger than 17uF, and most budget caps have a variance of +/- 10%, so in the grand scheme of things, it will mean slightly lower extension on the tweeter. It's more important than your caps are matched into pairs than if the value is +/- a few percent in one direction.

High-end caps typically offer +/- 5% variance so an 18uF can be as low as 17.10uF or as high as 18.90uF soif you have the opportunity, get them matched into pairs.

Badd99 · « **Reply #9 on:** 7 Jan 2023, 10:28 pm »

Quote from: ketchup on 7 Jan 2023, 07:19 pm

It looks like my speakers only have 17uF at the tweeter (there is no 0.1uF cap). In theory, how will 18uF in this location effect the crossover? Are there any other premium caps that will work well in this location that are 17uF?

As noted, your current caps have likely +/- 10% so using 18 here is of no concern. I'd add the duelund pure copper bypass cap 0.01uf with this. Will give you incredible improvements. Worth every cent.... Get them all matched at 1% (option when buying the vcap in matched pairs.)

JCarney · « **Reply #10 on:** 8 Jan 2023, 12:36 am »

Danny/Hobbs correct me if I am wrong, but don't the DCR and inductance of new inductors need to match the current inductors? Something that was brought to my attention re-building my Martin Logan Scenario's. I was lucky in that the inductors used were Erse air core, so I left mine in place.

JCarney

Hobbsmeerkat · « **Reply #11 on:** 8 Jan 2023, 01:44 am »

Quote from: JCarney on 8 Jan 2023, 12:36 am

Danny/Hobbs correct me if I am wrong, but don't the DCR and inductance of new inductors need to match the current inductors? Something that was brought to my attention re-building my Martin Logan Scenario's. I was lucky in that the inductors used were Erse air core, so I left mine in place.

JCarney

In practice it's mainly an issue for the woofer circuit, where you typically want similar or lower resistance than the original coil.

Air core inductors of the same gauge will certainly have a higher resistance than an equivalent gauge/value iron core, so you will typically need to increase the gauge to match the difference.

Ex: A large 16-gauge iron core (say 6mH & up) will need to be swapped for a 12 gauge air core or larger to get similar resistance.

Most speaker brands tend to use 18-22 gauge or smaller wire on their coils, so its easy to get away with 16 gauge air cores.

JCarney · « **Reply #12 on:** 8 Jan 2023, 01:53 am »

Thanks for the explanation Hobbs. I even understood what you were saying.

Best of luck with your project ketchup.

JCarney

Danny Richie · « **Reply #13 on:** 8 Jan 2023, 05:00 pm »

Quote from: JCarney on 8 Jan 2023, 12:36 am

Danny/Hobbs correct me if I am wrong, but don't the DCR and inductance of new inductors need to match the current inductors? Something that was brought to my attention re-building my Martin Logan Scenario's. I was lucky in that the inductors used were Erse air core, so I left mine in place.

JCarney

That is actually of very little significance considering that the bass and mid panels are in parallel and playing on top of each other. The bass panel has no filter and the mid panel has no high pass filter. It only has a filter pulling off some of the top end. So the impedance is going to drop down into the 2 ohm range (or lower). There will also be a lot of comb filtering in the horizontal plane. These things need a redesign real badly.

JCarney · « **Reply #14 on:** 8 Jan 2023, 08:56 pm »

Thanks Danny, I appreciate the explanation. I'll get out of the way now so ketchup can continue without my interruptions.

JCarney

ketchup · « **Reply #15 on:** 8 Jan 2023, 09:54 pm »

Quote from: JCarney on 8 Jan 2023, 08:56 pm

Thanks Danny, I appreciate the explanation. I'll get out of the way now so ketchup can continue without my interruptions.

JCarney

No no, it's all good! I'm getting pretty pumped for this project. These things sound so good biamped with a line level crossover on the lows/mids already.

JCarney · « **Reply #16 on:** 9 Jan 2023, 12:17 am »

Pics, take lots of pics cuz if there aint no picks, it did not happen.

Good luck and have fun,
JCarney

ketchup · « **Reply #17 on:** 10 Jan 2023, 02:30 pm »

Quick noob question about this network. Why is the negative side of the mid driver connected to the positive input? Is the schematic just showing that that portion of the crossover inverts phase?

Hobbsmeerkat · « **Reply #18 on:** 10 Jan 2023, 04:17 pm »

In every Maggie we've looked at, the tweeter/mid portion of the main panel is always wired with reverse polarity to maintain the acoustic phase with to the bass panel.

ketchup · « **Reply #19 on:** 10 Jan 2023, 04:27 pm »

Quote from: Hobbsmeerkat on 10 Jan 2023, 04:17 pm

In every Maggie we've looked at, the tweeter/mid portion of the main panel is always wired with reverse polarity to maintain the acoustic phase with to the bass panel.

Thanks, but I'm specifically talking about the mid and tweeter only. Why is the mid driver - at the top (blue wire) if it's connected to the positive input? The top of the tweeter is connected to the same input, and it's +.