advertisement


PFM Special - a simpler and more affordable DIY loudspeaker design

James said:
They go up to 22uF, but Mike needs only a pair each of 3.9uF and 12uF.
Click to expand...
You commented on 3.9µF, but the docs I have from Garf show 4.0µF. Which do I need?

Can you see any other issues in the docs that I pasted in the prior page?

Also, some of my inductors are 16AWG rather than 18AWG, which could have a lower series resistance than quoted. If I measure it and it's low, should I add another resistor inline?
 
Mike Hanson said:
You commented on 3.9µF, but the docs I have from Garf show 4.0µF. Which do I need?
Click to expand...
Jantzen does not make 4uF Z-Superior caps. 3.9uF is close enough.

Mike Hanson said:
Can you see any other issues in the docs that I pasted in the prior page?
Click to expand...
I don't think so.

Mike Hanson said:
Also, some of my inductors are 16AWG rather than 18AWG, which could have a lower series resistance than quoted. If I measure it and it's low, should I add another resistor inline?
Click to expand...
No, again unlikely to be material. If the 16AWG coil is used on the woofer filter, it will marginally decrease Qes which might have the effect of leaning out the LF in theory. But I don't think you will hear it practically. Just position to wall boundaries to suit your taste.
 
I'm mulling over the crossover. A few things to note, starting with the potential layout for the woofer and tweeter portions:

Crossover W&T.jpg

  • I expect I'll need a few jumper wires, given the tight quarters.
  • Is it OK for things to be packed in this tightly?
  • I was thinking to have the segments spread across the board left to right in frequency order: WMT. There's a bit of a gotcha, though, so I'll probably shift the tweeter section to the middle, and here's why...
  • Look at the size of that inductor for the woofer! :eek: The specs call the an iron core inductor in the woofer and midrange circuits, and this is what I've got. Looking at the pictures from the original prototype (on the prior page), all four inductors look more like air-core, although I don't know how they could be that small given the values. (Note that the iron core on the left is about 10x the value of the air core on the right.)
  • I believe the tweeter's inductor is properly oriented not to interact with the woofer's inductor.
  • Given this larger size of these two inductors, my board real estate is significantly restricted. I think I can get it all to fit. First assume that the tweeter section is shifted to the left, up against the woofer section. I might flip the woofer section end-for-end.
Here's the midrange stuff (not so tidy):

Crossover M.jpg

  • There's the other massive inductor. That's as far as I can separate the two big iron core inductors, so hopefully that's enough space between them. Unfortunately, they must in the same horizontal plane. As you can see, variations don't improve things (much):
    • I might flip one end-for end, which would make them a tad further apart.
    • Pivoting one horizontally 90° would make them closer, and their flux would still be in the same plane.
    • Turning one on end won't work, because the cubby isn't tall enough, and there's no way to secure it in that position.
  • There's a second inductor in this circuit, and I believe the orientation is correct so that it doesn't interact with the first inductor. However, it's in the same orientation as the tweeter's inductor, but is on the opposite edge of the board, hopefully far enough away to minimize interactions.
  • Note that I have a mix of Solen(Mundorf) low-inductance resistors and fancy Mills non-inductive. The Mills that's shown to the right of the inductor is the one that could have different values depending on my taste. Kludge something on that side so that I can swap off resistors fairly quickly to audition the different values.
As far as those Mills and Solen(Mundorf) resistors, I understand that the reduced inductivity is preferably for higher frequencies, which is probably why James indicated that the (much more inductive) sandcast resistor is fine in the woofer circuit. I've got a few extra Mills resistors on hand, so I could potentially use more in my circuits. They also are smaller, so easier to fit in here and there. Is there a particular focus as to where their would be best used?
  • I assume they're more useful in high-frequency settings, right?
  • Am I also correct in assuming that inline resistors are more important than those in parallel with the driver?
Any other thoughts, now that you can see my madness?
 
Mike, both L1011 and L2041 are iron-core inductors, but with 18AWG wire instead of 16AWG. Can you get them swapped?
 
James said:
Mike, both L1011 and L2041 are iron-core inductors, but with 18AWG wire instead of 16AWG. Can you get them swapped?
Click to expand...
That's what I was mentioning above, with my concern about their DCR. :oops: I ordered these parts years ago, so exchanging them is not an option. I would have to buy something new, which I'm willing to do.

However, I've checked Madisound, Parts-Express and Solen, and I can't find anything suitable. The options were rather limited back then as well, which is why I ended up with the correct inductance, while mistakenly assuming that getting a larger gauge would be no biggie.

Could I just add a resistor in series?
 
The reduced DCR is not a problem, and I don't recommend adding a resistor to bring it up to spec for the reasons I outlined earlier. I was trying to help resolve your space issue. I'll have another look for suppliers later.
 
James said:
The reduced DCR is not a problem, and I don't recommend adding a resistor to bring it up to spec for the reasons I outlined earlier. I was trying to help resolve your space issue. I'll have another look for suppliers later.
Click to expand...
Ah, I see what you mean. I do think I can get everything to fit, but it will be tight.

Alternatively, I could make two boards for each speaker, and mount them on either side of the cubby, leaving space for connectors and wires in the middle. The sides are 174x130mm, so easily enough room for the woofer and tweeter on one, and bandpass for the midrange on the other. That would also keep those big iron core inductors further from each other.
 
Just did the calculations, and splitting from the bottom to two sides doesn't give me TONS of extra room. The combined sides are about 7% larger than the bottom, but does provide the benefit of having the inductors further apart.

Yes another alternative: determine whether which needs more space, the woofer+tweeter or the midrange. Put the larger board on the bottom, and put the other on the side. That give the added benefit of the big inductor pivoting into another plane.

So I have options.
 
I'd be careful to secure the heavy inductors, if installed against gravity. You'll need more than double-sided tape.
 
Project update:
  • I've just picked up my Birch ply, cut into panels, so I could start assembly. I'll have to see if I can get into the shop tomorrow.
    • I have to design the layout and test the CNC router to cut the holes in the four baffle boards. This is a new skill to learn, hence the need for testing.
      • The top and bottom inner baffles will have simple circles, with the woofer and midrange holes rounded on the back side to minimize turbulence.
      • The top and bottom outer baffles will have stepped circles, so the speakers sit flush in the baffle. I think I'll be adding veneer later (at least to the fronts), so I'll mount them 2-3mm proud of the holes.
    • In case I need to wait for group learning on the CNC router, I can use the biscuit joiner to start assembling the inner box. (The last step is to apply the inner baffle from the prior step.)
  • I ordered from some of the fancy Jantzen Superior Z-Caps (12µF for C2031; and 3.9µF for C3011), and a couple of Jantzen Superres resistors (7.5Ω for R2211) that I accidentally omitted from an earlier order. They should show up next week.
  • I ordered some crossover terminals and a possible alternate 1Ω resistor for R2071 that's inline with the midrange.
  • I could start assembling the crossovers for the bass drivers, but I'll likely wait on those components above, as the larger caps may impact the layout.
 
Last edited:
Good Morning All,

As I read this thread I cast my eye to the thread title - simpler and more affordable??? - I guess all things are relative?

Regards

Richard
 
Richard Lines said:
Good Morning All,

As I read this thread I cast my eye to the thread title - simpler and more affordable??? - I guess all things are relative?
Click to expand...
Let's call it Optionally Affordable. ;)

My hope is that by focusing on good components, I'll boost the performance. And I would like them to look nice too.
 
You must log in or register to reply here.


advertisement


Back
Top