Paco's 1.6 Modifications
Originally posted: July 2000
Updated: December 2000 (every addition in bold type)
Further upgrade: April 2001 (every addition in red)
Still more upgrades: February 2002 (every addition in green)

Maggies 1.6 are wonderful speakers out of the box, and even much better after a couple of months of heavy use, but they can be easily transformed with a few, easy and inexpensive mods, until converting them into true impressive and even scaring sonic devices, perfectly able to compete with more, much more expensive speakers (and even beat them).

I have incorporated the following mods, with the help of many people in the AA and the MUG boards, so thanks everybody. I've been using my Maggies with a few unlisted tweaks, which also additionally improved the sound delivered (eg: tilting forward a little, spiking them on heavy granite blocks and using sand to make them more rigid). To see more about those tweaks you can easily search the MUG archives, and also take a look at my system for reference. Also, here is some pics of the system. Here I'll concentrate in the following:


CONTENT

O. GENERAL SET UP
O.1. THE CARDAS METHOD
O.2. TWEETERS ON THE OUTSIDE
O.3. IMPROVING RIGIDITY
O.4. TO SUB OR NOT TO SUB

I. SOME BYPASSES
I.1. BYPASSING THE CONNECTION PLATE & FUSE
I.2. STEPS SAVED
I.3. BYPASSING JUST THE FUSE
I.4. BYPASSING THE PANELS PUSH TABS

II. TAKING THE SOCKS OFF

III. CROSSOVER UPGRADE
III.1. REWIRING XOVER NETWORK
III.2. CAPS REPLACEMENT
III.3. INDUCTOR REPLACEMENT
III.4. TOTAL COST
III.5. FURTHER UPGRADE PATHS

IV. BIAMPING

If you happen to try any of these mods, please post your results on the MUG board. Also, I'll be happy to provide you with more info or help as long as I can.


O. GENERAL SET UP

O.1. THE CARDAS METHOD. In my experience, the Cardas method is the best for my Maggies, especially when they form an equilateral triangle with my head (nearfield). For details, and several possibilities, see here. A lot of experimentation is indispensable, though.

O.2. TWEETERS ON THE OUTSIDE. After some time with the tweeters on the outside, then on the inside, I now think the tweeters on the outside sound much better, in the long run. See what I wrote about this. You can also experiment with taming the rear tweeter beam with something; compact plants did the trick for me, although after biamping I rather put them on the corners.

O.3. IMPROVING RIGIDITY. The problem is the Maggie feet are not enough to make the panel even moderately rigid. Among the several possibilities available I chose to add aluminium bars on the rear side. After a first attempt with just one bar, I then added a second. For details see here

O.4. TO SUB OR NOT TO SUB. No doubts here: a good sub adds what is needed in terms of bass, presence, weight and more. I’m very happy with my Rel Strata III, as you can read here.


I. SOME BYPASSES

Bypassing usually means making the signal path shorter, then obtaining more transparency; that's why I made the following ones to my Maggies, but there is no need to do them if all you want to do is rewiring the xovers or replacing the caps and/or the inductor; also, to do these mods you don't need to take the socks off.

I.1. BYPASSING THE CONNECTION PLATE & FUSE

I was fed up of the cheap connection plate of my 1.6, so I decided to bypass it all, including the fuse; to this end, you can do this (it will take you about 1 h each Maggie):

  1. unscrew the plate;
  2. use the soldering iron to release the wires from the binding posts (even better: cut them);
  3. join red to yellow (+) and black to brown (-) and you are ready to connect the 1.6 again;
  4. if you use speaker cables, you can connect the wires to them, wire to wire, to further benefits;
  5. in case you use monoblocks, you can put them very close to each Maggie, then connect the Maggies wires to the amp binding posts (even better: open the monoblocks, release the amp output wires from the binding posts, unscrew them, take the wires out through the holes, then connect them to the Maggies wires, either by strongly clamping them together or by soldering them).

How much transparency can be so obtained? You can measure it by counting the steps you are saving.

I.2. STEPS SAVED

How many steps there are between the amp output bare wire and the speaker input bare wire (the one from the speaker binding posts to the xover input)? Let's see:

AMP/SPEAKER CABLE: solder/binding post/contact banana-binding post/banana/solder/: 5; total (+ & -) 5 x 2 = 10.

SPEAKER CABLE: cable/solder/banana/contact banana-binding post: 4; total (in biwiring mode): 4 x 4 = 16.

SPEAKER (Maggie 1.6, biwired): (i) binding post/solder: 2; total 2 x 4 = 8; (ii) (fuse wire): wire to fuse box/solder/fuse box/contact fuse/fuse/contact fuse/fuse box/solder: 8; total each speaker: 8 + 8 = 16.

Total amp+cable+speaker: 10 + 16 + 16 = 42 steps; there are TWO speakers, so the FINAL TOTAL is 42 x 2 = 84 STEPS!!!

So in the very moment I finally CLAMPED the output bare wires of my monoblocks to the input bare wires of my Maggies, so bypassing binding posts, speaker cables, fuses, etc., etc., I was saving 84 steps to the signal path.

Results? Transparency, transparency, and still more transparency; but you don't have to believe me, you can try yourself, and you'll probably be amazed, as I was myself.

I.3. BYPASSING JUST THE FUSE

In my opinion the best way to bypass just the fuse is this:

  1. unscrew the connection plate;
  2. look for the xover wire that is soldered to the fuse box;
  3. release it with a soldering iron (or just cut it);
  4. resolder it to the binding post at the other end of the fuse;
  5. return the plate to its place.
In this way you replace this situation: [wire/solder/fuse box/fuse contact/fuse/fuse contact/fuse box/solder/wire/solder/binding post] by this one: [wire/solder/binding post], so saving EIGHT steps to the signal path, and not only the fuse itself, which would be the case if you just replace the fuse by some quality wire. The result is a noticeable improvement in transparency, but if you bypass the whole plate the improvement is clearly superior.

I.4. BYPASSING THE PANELS PUSH TABS

This is a very important tweak to improve transparency and musicality, and can be done together with any other xover tweak, either internal or external; when bianping I think this is a must, and very, very easy. See my results here.


II. TAKING THE SOCKS OFF

In order to have fast and convenient access to the stock xovers (and also to enjoy a more transparent sound, with no veils) I took the socks off. It took me about 90 minutes each Maggie, and the main points to be carefully considered are these:

  1. On removing the trims, you have to be very careful not to break them, as the only way to do it seems to be by introducing a flat screwdriver between the trim and the Maggie and pull it out, step by step.
  2. Most of the nails remain on the Maggie, but some may remain on the trim, so it is important to label each of the four trims, including the position (up, down), in case you decide to reinsert them to their original places.
  3. Removing the staples is easy, after unscrewing the stands, but be careful not to damage the fabric; the same flat screwdriver can be used to good results; one pin of each staple out is enough: after that, you can use the pliers; there are two rows of staples, over 20 each.
  4. On taking the socks off, first thing to do is unscrewing the connectors/fuse plate (there is a hole in the fabric); the main concern is to avoid that the fabric be damaged by the heads of the nails which have remained on their original places; don't forget the isolated staples which are located just on the top, each side.
  5. If you want to tweak the xover, a further, stronger piece of fabric should be took off, but it is small and is simply glued to the wood; also, this same piece is protecting the connections of the wires to the panels.

According to Magnepan, the diaphragm and adhesive of the new models are UV resistant, and this is very good news, especially for me, for after hearing my 1.6 without the socks I've decided to have them permanently naked (they look uglier, but definitely sound better).


III. CROSSOVER UPGRADE

Everything is a compromise, and I'm taking my own choices here, being my ABSOLUTE priority not to add unnecessary steps to the signal path (including speaker cables), as for instance binding posts, connectors, solder, several runs of wire which can be saved, etc., etc., all involved in building an external xover box.

Also, I have to say that IMO many people won't cut the fabric nor build an external xover because the first involves a strong aggression against their beloved Maggies, while the second is expensive and time consuming; yet you can have an excellent xover upgrade without having to cut the fabric nor paying a lot of money, with everything located in the original space available in your 1.6.

To do this, you don't need any previous technical training or special skills, as I didn't have such things myself.

III.1. REWIRING XOVER NETWORK

The stock xover network wires are not good, to say the least. There are 6' of 16ga cheap tinned copper, cheap PVC insulated, with lots of unnecessary solder tabs. I made a first experiment replacing them by:

This resulted in clear improvements, but I was not quite satisfied, so I decided to add some silver (I think Maggies like silver) like this:

(I ordered the wires from
homegrown audio) Some more details:

Results: more transparency and low level detail; more resolution and separation; a noticeable noise floor reduction (I suspect the van den Hul speaker stranded hybrid wires were doing more harm than good); better bass; timbres improved, especially brass, but also strings and percussion.

(At present I’m using almost no wire, as I’m connecting directly each end of the inductors and capacitors to the output wire of my mono amps and to the push tabs of the Maggie panels; for the negative end of the panels i’m using solid core, enameled 12ga copper for woofer and 15ga for tweeter.)

(NOTE ON CLAMPING WIRES. For those of you who may be interested, I always prefer clamping 2 wires together than soldering them. To do that, I use small metallic cylinders (generally brass), plastic insulated, with two small screws on each side, which are perpendicular to the cylinder axis. There come in virtually every size, and are commonly used by electricians, at least in Spain. So you put the wires to be connected (two or more of them) into a cylinder of the suitable diameter (the lower the better), then simply tighten the screws as much as you want, with the only caution not to break the wires. Notice that the connection is good because you have the wires strongly pushed to each other, so you don't depend on the quality of the brass. I'm using them almost everywhere, including mains connections, to very good results. They are great, and of course all connections can be easily reversed. See more on these screw/crimp connectors here, and a photo here)

III.2. CAPS REPLACEMENT

After some doubts, I finally chose Ansar Supersound polypropylene axial capacitors for the replacement. Ansar caps are highly regarded among audiophiles, especially in Europe, where they are well-known, and are used by some top-notch speaker companies (Proac is a good instance).

My Ansar caps are just one 22uF and one 25uF for each Maggie, and just 5x4 cm size, which allowed me to install them on the same place of my former stock caps, with no need of an external xover, so avoiding additional wiring and connections, as well as the smearing time and microphony problems involved in using several caps instead of one for each value; besides, there is room left for replacing the inductor for a better one. Also, they are 400v, for high voltage caps tend to sound better.

After a few days of almost continuous playing, the main sonic results of replacing the stock caps by Ansar ones were these: much more low level detail, a much fuller (although not particularly bigger) soundstage, and more natural timbres for almost all instruments (including human voices); all kinds of music now sound more transparent, with more immediacy and cohesiveness.

This MAJOR upgrade recalls me the replacement of my former preamp NOS 7308 Siemens and Telefunken tubes by my present Amperex, but just for a fraction of the money I paid for them.

The price I paid for my Ansar caps was $12 each. The Ansar web site is here. There you can see some info about distributors. By that time Wilmslow Audio had the 25uF cap in stock, but not the 22uF, so i ordered them from Cricklewood Electronics. I don't know the US Ansar distributor, but you can ask Ansar by email.

You can contact Wilmslow by email and Cricklewood by phone (+44 208 452 0161) or fax (+44 208 208 1441). Also, you can see good Cricklewood ads in the UK magazine Hi-Fi World. (For Cricklewood’s website see here; in case the site is not working, their email address is this sales@cricklewoodelectronics.com. If you happen to order from either of them, make sure they will send you exactly what you order, with no replacements.)

Summing up: this is the most cost effective mod i've ever made.

(For details about my experiences with different caps, see below.)

III.3. INDUCTOR REPLACEMENT

The criteria for my choice were these:

  1. to match the same original space available (so avoiding an external xover, with its unavoidable additional wiring and connectors);
  2. to respect Magnepan's explicit recommendation not to replace the stock inductor by an air core foil type (to see more details about that recommendation, you can read this message
  3. to get as low DCR as possible (0.4 ohms the stock one), while keeping the same inductance (3.5 mH), which is usually reached through increasing the gauge (19 ga the stock one).
After some unsuccessful attempts to have such an inductor custom made, I found that madisound.com has been using and selling what they call "Sledgehammer steel laminate inductor", which in addition to meet my needed specs (3.5mH, 15ga, 0.19 ohms DCR, 95x45x32 mm), is very inexpensive (about $12 for my value); you can see the info here. These inductors are very similar to the 1.6 stock ones, being the size and gauge the only apparent differences, and as it seems they are being used for xover networks in much more expensive speakers.

Results: very nice indeed, although nothing comparable to the ones obtained through the caps replacement; in particular, the bass is now cleaner, faster and louder, with more visceral impact, which is clearly noticeable with all sorts of music where bass is important; better bass is particularly noticeable in rather boomy recordings, which are now much more enjoyable

To get not only better bass, but MORE bass is specially nice to me, as I always found the tonal balance of my 1.6 rather biased to midrange and treble; now I enjoy big symphonic orchestras, big operas and big choral works much more, but also my acoustic jazz favourites.

If you happen to share my above three criteria, or think that the tonal balance of the 1.6 should be slightly corrected towards more bass, or simply want to experiment for just $25, then go for it; it is highly recommended.

(NOTE ON STEEL LAMINATE INDUCTORS. Not having expertise on inductors, I looked for some advice in the industry. This is what Stephen Tidwell, the owner of Layne Audio, wrote to me recently about steel laminate inductors: "Regular iron bar or rod cores [also called ferrite] are made up of a low density "powdered" iron material that is cast into a given shape, be it a square bar, round rod, or more complex shape like a bobbin. Steel core inductors use a plain mild steel rod or bar. Steel laminate inductors use a bar made up of many thin layers of grain oriented steel alloy sheeting, much like the construction used for certain types of transformers. As I understand it, the advantage comes from the density and grain structure of the steel alloy layers. The core is easily and evenly permeated by the magnetic field generated in the windings, which leads to very low distortion at lower frequencies compared to even air core inductors. Iron /ferrite cores have comparatively poor magnetic qualities, hence the added distortion.")

(NOTE ON FOIL INDUCTORS. Foil inductors are HUGE (the 3.5mH, 12ga is about FIVE pounds of copper!!), and their main goal is to prevent against the "skin effect": a distortion and loss of power taking place when HIGH FREQUENCIES pass through thick conductors. Yet the 1.6 xover point is 600hz, which is an impressively LOW point for a two way speaker, so the inductor main mission here is precisely to KILL high frequencies, not to make them happier. Some speaker builders, even though they do admit that foil inductors may work very well, they still prefer not to use them above 2.0mH. Also, foil inductors exhibit some distortion too, and there are serious people in the industry thinking that a well implemented steel laminate inductor may exhibit even less distortion than foil inductors, especially above 2.0mH inductance. Finally, foil inductors are very expensive.)

(For details about my experiences with different inductors, see below.)

III.4. TOTAL COST

The total cost of my xover upgrade for the two Maggies is $85 ($12 wires + $48 caps + $25 inductors).

(These prices are considerably increased with Auricaps/Jensen caps and North Creek inductors; see below.)

III.5. FURTHER UPGRADE PATHS

Caps. I may try Auricaps in the future; they make the 22uF exact value, 200v, 1.67x1.75, at $38; you can get more info here (I tried the Auricaps, as well as the Jensens –about same price as Auricaps, so you can see a full comparison, including a good Jensen source, here and more info on the Jensen line here.)

Inductor. A further replacement might be the air core12ga, 85x44mm, 3.5mH, 0.30 ohms DCR, made by North Creek Audio; their inductors are prestigious and well built, but you would have to pay about $40 each; you can get more info here. (I tried the North Creek inductors as well; for details about my sonic results you can read here.)


IV. BIAMPING

Instead of boring the readers with the full, detailed story of my biamping adventures and results, here you have all the steps as they appeared on the AA boards

Jan 9: first intention to biamp: noted here

Jan 25: passive biamping (still with high level xover), first impressions: noted here

The line level xover circuit used (designed by Davey explicitly for my amplifier's input impedances : THANKS A LOT DAVEY!!!!!!):

Feb 4: first impressions at inserting the passive line xover: it works!!: noted here

Feb 8: update: noted hre

Feb 11: darkness problem solved noted here

Feb 15: components upgrade (Rel caps & Vishay resistors)

March 20: upgrading the tweeter cap: Jensen did it again

April 16: upgrading the tweeter resistor: tantalum vs Vishay

xover update (Feb 11, 2002): better tonal balance


MUG Home