• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Can anyone help with Magnepan 3.5r external crossover build?

Brandosees

New Member
Hello, all.
I am interested in assembling crossovers for my magnepan 3.5r’s. I have no prior experience in Building a crossover.
I was told that my post was not specific enough. I will try to correct that now.
I was wondering if anyone with knowledge about this would not mind sharing ideas
I am having difficulty finding the correct values for the inductor coils.
I have found several which would work, but I wish to ask certain questions such as how various parts may alter the sound.
I’d like to adhere to the original values in the magnepan schematic, which I found. I’d like to build it close to the way they did, but using superior parts.
Can anyone who has experience with upgrading/building crossovers, assist?
Thank you, kindly.
 
Last edited:
Register to hide this ad
Hello, all.
I am interested in assembling crossovers for my magnepan 3.5r’s. I have no prior experience. I was wondering if anyone with knowledge about this would not mind sharing ideas? Thoughts? Parts list?
How to find the correct values?
I am having difficulty finding the correct values for the inductor coils.
I’d like to adhere to the original values in the magnepan schematic, which I found.
Can anyone assist?
I'm not an EE, nor do I have "vast knowledge" in any sense of those words. But I do have some DIY experience in building crossover boards. First, a question:
By the "original values for the inductor coils" do you mean the Bass and MR/TWT crossover networks dated 8/31/93? It is essentially the same design as the 7/12/95 version, shown below it, with some minor differences.

The 8/31/93 version shows the inductor coil values with 3-digits, such as 8.52 mH. You'll never find inductor coils wound that precisely, nor do you need them. Instead of 8.52 mH, 8.5 will do just as well. Same thing goes for the other inductors.

The schematic also shows two 2.86 mH inductors linked in series. This was done to achieve a total inductance of 5.72 mH. This was likely done because they didn't have a 5.72 mH inductor available. In audio crossovers, it is generally advised to use a single inductor instead of linking two inductors in series. I'm not completely certain why this is, but it probably is done to avoid having the electromagnetic fields of the inductors interfere with each other. See the thumbnail for a general rule to arrange multiple inductors on a crossover board. If you link two inductors in series, and if you make sure that the two inductors are arranged as in example 6, you should be alright. As you can imagine, with many inductors, keeping them 4" (10 cm) apart on a board becomes a big problem.
upload_2022-3-19_12-15-20.png
That original 8/31/93 schematic also shows the inductor coils with symbols than can mean either generic inductors or air core inductors. The 7/12/95 version specifies those inductors as iron or laminated steel core inductors. I think you should use the 7/12/95 version schematic instead of the original version, here's why: Air core inductors use much more copper wire than the iron/steel core inductors. As a result, they have much greater DC resistance, that will significantly reduce the bass signal voltage (known as insertion loss), with a noticeable decrease in sound quality. They also cost quite a lot more. Air core inductor electromagnetic fields are less likely to become saturated than iron/steel core inductors. For this reason, they seem to be preferred among some audiophiles, but those large air core inductors are significantly more expensive, as you've noticed. If you choose inductors with large gauge wire, and laminated steel cores (not iron cores), such as these inductors, they can work well for your bass networks. (Madisound is a good source for inductors, steel laminate or air core.)

You will probably not be able to find inductors with exactly the values shown in the crossover design. But there is an easy way around this. Buy a pre-made inductor with larger inductance than what is shown in the schematic. For that very large 8.5 mH inductor, buy this 12 mH one from Madisound. Also, buy an inexpensive LC or LCR meter, such as this one from Amazon. Unwind wire from that larger inductor, but don't cut it off yet. Use the meter to measure the inductance. As you unwind wire, the inductance gets lower. When you've unwound enough wire to get 8.5 mH, cut off most of the excess wire. Leave about 6-8" of unwound wire, so you can easily assemble and solder things later.

Finally, you might have to phone Madisound to ask for custom wound inductor coils.
 
Last edited:
Thank you. I had already sent him a message, but I don’t think he can tell me much as he went for a very inexpensive build. I am looking to upgrade.
But I appreciate the reply.
would have been appropriate to state you wanted to upgrade in your original post. That first post said you were looking for help making a set of crossovers.

Buy the good stuff, you should be fine.
 
A question to the OP... What don't you like about the existing crossover? What are you trying to achieve?

Lastly, here is a cool build I found online. In this one, they also included the internal crossover. I do not wish to do that. I’d like to do something along these lines, only with the external crossovers and leaving the internals alone.
I would have contacted the user who posted this on the maggie forum, but the post was many years old. If anyone could and would assist, I would greatly appreciate it.
Thank you.

What attracts you to this "cool" build?

Huh..JPG
To me, that looks like someone has WAY more money than sense. I've never seen a high quality crossover with the coils oriented like that. It might look cool, but typically coils are oriented so that they lie in different planes, and/or are separated ....
https://www.audioholics.com/loudspeaker-design/inductor-coil-crosstalk-basics

If your current crossovers are malfunctioning, I'd contact Magnepan, or just rebuild them as-is using quality components with specs close to the original components in your crossover, using the schematics as a backup. If they are not, and this is more of a "I'd like to do a fun $500 audio project," .... well, there's lots of fun things to build rather than replace something that you already own that works really well. Many times, "upgrades" don't end up being upgrades. The original designers of quality gear really do know what they're doing, and Magnepan falls in that category. I love my 3.6Rs exactly the way the factory built them, and I'm not planning on adulterating them in any fashion.
 
I was just asking what your motivation was; it wasn't clear in your original posts. Additionally, the picture of the coils all side by side in close proximity like that worried me and I didn't want you to make that mistake. I've built a lot of speakers and crossovers, some my own design, some others, some from kits. and was just trying to help. To do that, I needed to know what you were trying to do. If you had just said, "I want to rebuild my factory crossovers as designed, just using better quality components," then we would have better known how to help you, although you did say that you would "like to adhere to the original values in the magnepan schematic." Given that you said you have no prior experience, I didn't know if you meant the same crossover points, slopes/Qs, or the same component values.

In any case, I still would have asked what you don't like about what you currently hear, and what you'd like to improve. That's a valid question that helps focus in on the goal. You led with " I have no prior experience. I was wondering if anyone with knowledge about this would not mind sharing ideas? Thoughts? Parts list?" I have some knowedge, some prior experience, and I didn't mind sharing some thoughts and questions. I thought that's what you wanted. My approach to audio is, and always has been, to identify areas of deficiency in sound quality and/or reliability, then address them. That's all.

R. Swerdlow gave you some good thoughts. I tried to, too, but evidently they weren't the thoughts you had in mind. That's OK.
 
I agree with MuseChaser when he said this:
I've never seen a high quality crossover with the coils oriented like that. It might look cool, but typically coils are oriented so that they lie in different planes, and/or are separated ....
No matter which inductors you choose, never arrange them as in that so-called 'cool looking' photo. There will be significant crosstalk between their magnetic fields, causing their inductance values to change from their printed values. See that diagram I included in my earlier post. Examples 1, 2 or 3 correspond to what that photo shows. I can't tell the dimensions in the photo, but I doubt if those large coils have center-to-center spacing of 20 cm.

Also see this: http://www.troelsgravesen.dk/coils.htm

I also wouldn't recommend spending extra for those flat wire inductors. They offer little, if any, audible benefit over standard wire inductors. However, I do recommend spending extra for thicker wire gauge inductors to minimize insertion loss. Ultimately, it's your $ and your choice.
 
Last edited:
Your first message was not like this one. It was a lot of opinion with no information.
Had you said what you just did, I would have asked questions and had a discussion. The first message you sent was most unhelpful.
Now you’ve revised it and are projecting things onto me. you asked no specific questions related to the details of the build. Just about why I would even want to do it and then making clear that you frowned upon it.
I certainly did not come on here to argue.
If you don’t want to be of assistance, no problem.

The only opinion I offered was in the last sentence of my first post... " I love my 3.6Rs exactly the way the factory built them, and I'm not planning on adulterating them in any fashion." Everything else was either a question or a fact. I was trying to help. Trying to help you avoid placing coils as in the picture you described as cool. Trying to figure out what your goal was so I could be of further help. I projected nothing onto you. I suggested rebuilding the crossovers matching the values of the existing components, and using the schematic as a backup, rather than using the schematics as your first source. That is a helpful tactic, based upon experience. That's what you asked for.

I have NO idea what you are so upset about. I do want to be of assistance, and thought I was. In any case, welcome.
 
Three different crossover schematics are referenced in 2 different threads. But, all 3 schematics are the same. Each version just uses differing amounts of components to reach the same values. Personally, I'd follow the schematic posted by @Sgt.Pepsi. The components in that crossover should be easier and cheaper to find.

Also, in glancing at Parts Express, all the various inductors and capacitors can be found in the exact values needed, or close enough to be within tolerance. So, it shouldn't be difficult to source the parts for the mods.

Finally, I don't really understand only modding the External Filter? That's seems to address half the issue. A mod on the bass panel, without touching the mid and treble driver crossover, isn't logical to me. Modding the Internal Filter would most likely produce the biggest audible improvement. To me, the "cool build" crossover would be the model to follow (along with a better layout). Otherwise, I don't see the point?
 
Last edited:
2ABF244C-6A11-44B0-9A90-583F51CA9CC6.jpeg

Looking at the External Filter, there are 2 paralleled capacitor bundles. Personally, I'd build each bundle as a hybrid. Use all NPEs in each, but one. That last capacitor would be a poly cap. The Mid/High bundle could be electrolytic 22+22+22+22+15+12, with choosing a 12, 15, or 22 as a poly. The 2nd bundle could be three 68 NPEs, with a fourth 68 poly. Normally, I'd suggest an Audyn Cap Plus for the 22uF, but it's had a significant price jump at PE. So, consider a Jantzen Standard-Z as an alternate. For the 68, a Jantzen Crosscap would be good.

As to coils, I'm no expert. But, PE carries very good inductor brands. Most like air core inductors. Match inductance as closely as possible. The larger the gauge, the better, for higher power handling. Though, that means bigger and more expensive, too. Be aware, resistance also matters with an inductor. If possible, measure the resistance on the stock coils, and match that parameter as well.

Good luck.
 
Last edited:
Some research which might help is to look into gunned Magnepans on the asylum's MUG

https://www.audioasylum.com/forums/MUG/bbs.html

Member Peter Gunn used to hot rod the speakers and that included a high performance crossover. PG, I think his name is John, stopped doing this work a number of years ago and only did a few of the 3-series, preferring to do the smaller speakers and saying the results were better.

But if you want to research it there may be some good info in the pages of posts about gunned Magnepans.

I don't know if PG's website is still available but the way back machine might have it. Don't know if he told folks what he did but at one point he was willing to talk to folks and discuss crossover upgrades folks could do at home.
 
Yes. And about a month ago when I reached out to a pro about having these upgrades done, besides being scolded for wanting to adhere to Magnepan’s design as closely as possible, I was told that the use of these “cheap” metals was done for budget reasons and that no one would use these parts today.

I did not argue.

But this made me want to further investigate, as I figured the folks at Magnepan must have had some reason other than budget.

Why would they want to have their speakers sound worse?

I understand some upgrades are possible, but I didn’t think the engineers were simply clueless about what they were doing, either.

Difficult to be clueless and end up with such a beautiful sounding speaker.
I don’t understand what you meant when you said “cheap” metals in the inductors. Are you talking about air core vs. magnetic core inductors?

There is no simple YES or NO answer about air core vs. magnetic core inductors in audio crossovers. You will find those who claim you should NEVER use a magnetic core inductor in a loudspeaker, but never is a too strong a word. There are some situations where magnetic core inductors make sense. If done right, it should not compromise sound quality.

All inductors, air or magnetic core, are made with solid copper wire of various gauges from 20g to 14g. Solid wire is required as stranded wire is too flexible to keep its shape as it gets wound. The wire is coated with a thin layer of lacquer insulation. So, all inductors are made with the same copper wire. There is no difference in the metal that conducts the audio signal.

Air core inductors are made by winding copper wire around a plastic or wooden bobbin. Some of these (Janzen inductors) are sold after removing the bobbin. Magnetic core inductors are made with the same coated solid copper wire wound around a magnet made with an iron bar or laminated steel plates. It has long been known that a magnetic core wound with a given number of copper windings creates a higher inductance value than an air core inductor with the same number of windings.

Because magnetic core inductors use less copper wire, they are less expensive. In the case of your Magnepan speakers, the low-pass filter for the woofer section uses very large inductance values because the crossover frequency is roughly 200 Hz. That's very low for an analog filter. They require long lengths of copper wire. This becomes costly, and leads to audibly high insertion losses, ie. speakers become less sensitive because of the inductor wire's DC resistance.

A good way to avoid this problem is to use magnetic core inductors. Their down side is that their magnetic fields can in theory become saturated more easily than with air core inductors. If more potent laminated steel magnets in the core are used instead of solid iron bars, and if large gauge copper wire, such as 15g, is used, this theoretical problem doesn’t appear in practice with the signal levels found in recorded music. Although it is widely debated among audio amateurs, no one understands what magnetic core inductors might sound like in loudspeakers if they become saturated. Is it audible at all? I for one have never heard it. Certainly the speaker designers at Magnepan understood this when they chose magnetic core inductors in the woofer section as a way to lower cost – without compromising sound quality. I’m surprised the ‘pro’ you consulted didn’t understand this.

As long as magnetic core inductors are limited to the very large inductance values required for a low-pass filter operating below 400 Hz, you should be fine. As frequencies get higher into the mid-range, where our ears are most sensitive, the required inductance values get much smaller. These smaller inductors can be air core types without the high cost.

I hope that answers your questions about ‘cheap metals’.
 
Last edited:
If I may, there are several other questions I have. They are all related to understanding the reasons for the original crossovers and making informed decisions while upgrading parts.

With poly caps, I was going to go with Dayton Audio. There were PMPC and DMCP. The PMPC is priced higher and has copper leads. Is this an advantage? Or possibly a drawback, given what you mentioned earlier regarding copper.
There is no advantage or drawback, other than price, if you choose the PMPC or DMCP capacitors. They have the same construction (other than the leads) and the same sound quality.

When choosing capacitors, the most important thing is getting the correct capacitance value. For most all crossover parts (capacitors, inductors or resistors) a general rule-of-thumb says crossover part values that vary by about 10% are OK. So, as long as you actually get the capacitance as specified in the crossover schematic, or are within 10% of that value, you’re doing the best you can.

The Dayton Audio capacitors are metalized polypropylene (MPP) caps, made in Taiwan. Overall, they’re inexpensive, but they’ve been recognized for years as having very good quality. Among the audio parts that Parts Express sells, their house brand capacitors are perhaps the best known among DIY builders. These caps are sold in two price levels, one with 1% tolerance (the PMPC line), and the other with 5% tolerance (the DMPC line). As far as I can tell, there is no difference in these caps other than the tolerance. After they’re made, in one large manufacturing run, they go through several quality control inspections, including testing for capacitance. If a cap deviates from its intended capacitance by 1% or less, its sold marked with 1% tolerance (PMPC). If the capacitance deviates by more than 1%, but less than 5%, its sold at a lower price marked with 5% tolerance (DMPC). If it deviates by more than 5%, it isn’t sold in the USA by Parts Express under the Dayton Audio name.

I once had the opportunity to speak at length with Darren Kuzma, who at the time was the parts manager for Parts Express. He often came to DIY Speaker Builder meetings around the country. He was popular because he always brought along freebies to give away. He was highly experienced at DIY speaker building, crossover design, as well as the manufacturing of speaker parts of all kinds and price levels in China and Taiwan. We mainly spoke about the new (at the time) Dayton Reference Series aluminum cone drivers he developed and had manufactured in China under contract. But, when I asked him about Dayton caps, he did confirm what I described above about the Dayton MPP caps. He also told me they were manufactured in Taiwan by a well-known electronics manufacturer named Bennic. That was some time ago, and I don’t know for certain if Parts Express still sells capacitors made by Bennic under contract. However, these caps are still considered very good quality and their price makes them an excellent value.

Another advantage of owning an LCR meter, is that you can easily measure the capacitance of any capacitor you buy. Among the various Dayton Audio caps I’ve purchased, ALL of them measured, in my hands, as nearly exactly their labeled value. It didn’t matter if it was a 1% or 5% cap. After seeing that, I only buy the 5% caps.

Solen and Audyn Q4 caps are also very good quality and sell at slightly higher prices. If I can’t find a Dayton 5% cap in the size I’m looking for, I order either a Solen or an Audyn Q4. Their only possible disadvantage is that most of them have failure ratings of 400V instead the 250V of the Daytons. Being able to withstand higher voltages before failure, does make for a physically larger capacitor. However, it does not make for improved sound quality.
 
Regarding the build I thought looked very cool, is there any advantage to using the heavier gauge copper wire twisted, to connect everything?

As long as you use a good ground connection, with good wire, I doubt that you'd hear a difference in using a heavy solid wire. The twisted wire was probably used to make for a more stable platform to wire (solder) to.

Thank you for this information. I am just getting around to it.
May I ask some questions about the information you’ve provided here?

Sure, ask away! That's why we're here...! :)
 
Last edited:
Peter Gunn ( John) is still active and does share DIY info for those that want to upgrade their older Maggies. He recently provided me with some Jupiter paper and oil caps to update my 15 year old Magnestand MMG's. He will respond to an e-mail, so ask away!
 
Regarding the build I thought looked very cool, is there any advantage to using the heavier gauge copper wire twisted, to connect everything?
I agree with StimpyWan. That twisted solid copper makes for a pretty photo, but I doubt if there is an audible difference.

There is a practical limit to heavier gauge wire, solid or stranded, when assembling crossovers. As the gauge gets thicker, it becomes more difficult to solder things to it. The leads on capacitors and resistors are much thinner. My own preference on crossover boards & internal connections is to use wire no thicker than 16g. I once tried using 14g, but quickly realized it took so long to heat up that I damaged some caps.
 
I couldn’t tell you what he meant. He did repeat several times that the parts were cheap and that cheap ferrous metals were used. When I asked to have them made as Magnepan made them, he said NO ONE who is a professional would use electrolytic caps and that it was impossible to recreate what magnepan had done. I wont mention who said this, but I will say that he is well-known for doing these upgrades and I was disappointed to see how rigid he was regarding the build.
That story makes me wonder if this so-called pro made more profit from selling high priced capacitors, inductors, and wires than he did from the basic assembly. Sadly, the third party audio upgrade business has too many people like that.
 
Back
Top Bottom