Z19 Crossover

[Zonker92]

On a different note, it would be so cool for us noobs (hint hint), if someone with mad graphics skilz were to lay out a simple Z19 in the following format (the image below is a crossover from the awesome troelsgravesen site). It would make building it So. Easy.

 

[Dr. Beaker]

What's the consensus on using a Z19 with the 2235H?

I know JBL doesn't recommend crossing that woofer over any higher than 1000hz, but the graphs for it don't look too much worse 1000-1200hz than they do for 800-1000hz.

Will the Z19's 1.2K crossover point be pushing it? Should I use Zilch's 4430 crossover interpretation instead?

JBL crossed the 2235 at 1000 Hz on the 4430. The crossover is N3134. I think there are a few that have used it with the Z19 and like it but, I have not seen measurements posted. It won't hurt the woofer to try it.

 

[ricohman]

Great idea Zonker.
That is something that should be done.
Zilch kept posting info for us long after he knew about his health problems and that takes courage. I still feel bad about bugging him for info, but I didn't know.:sigh:
I think he would have wanted the thread to have such a drawing. He said he wanted us to do the best we could with the skills we have. He was all about home-grown builds and projects.

 

[Sam Cogley]

I would love to have met the man, he seems to have been a genuinely great person. I need to get off of my butt and build my Z19s soon...

 

[srbliss]

Z-19 and the 2235

I discussed this with Zilch last year and he did not know - said try it out and measure the results - what a surprise. Without the mass ring I think it would be fine. I really want to try Gordons 2236 hybrid with this crossover.

What's the consensus on using a Z19 with the 2235H?

I know JBL doesn't recommend crossing that woofer over any higher than 1000hz, but the graphs for it don't look too much worse 1000-1200hz than they do for 800-1000hz.

Will the Z19's 1.2K crossover point be pushing it? Should I use Zilch's 4430 crossover interpretation instead?

 

[srbliss]

He wanted it that way

The Z-man was insistant that any of us that knew kept our mouths shut as he just didn't want people knowing about his problems and feeling sorry, or worring about him, said it wouldn't do any good for him or others so why bother. I don't think he told his remote family until near the end if at all. He was a very private man and I was lucky to have met him in person several times at ZilchLab.
Steve

Great idea Zonker.
That is something that should be done.
Zilch kept posting info for us long after he knew about his health problems and that takes courage. I still feel bad about bugging him for info, but I didn't know.:sigh:
I think he would have wanted the thread to have such a drawing. He said he wanted us to do the best we could with the skills we have. He was all about home-grown builds and projects.

 

[winters860]

I discussed this with Zilch last year and he did not know - said try it out and measure the results - what a surprise. Without the mass ring I think it would be fine. I really want to try Gordons 2236 hybrid with this crossover.

Gordon's 2236 would be great, but not an option from my E140 starting point.

I've been looking at this thread over on LH, and I'm not sure there's much advantage to leaving off the mass ring and making a 2234H instead of a 2235H. You'd lose a few db from the low end, get a bit better mid-band efficiency, and wind up with a very similar curve around the crossover. Unless the marginally better efficiency is integral to your design, I can't see a reason to go without the mass ring. With it, you get better bass extension and a slightly flatter response curve. Am I missing something?

 

[srbliss]

2234h

That was a comment from Zilch when he was measuring my speaker. Said he didn't know if the 2235 would be happy up to 1200 and suggesting that I make a Z-19 and try in out.
Steve

Gordon's 2236 would be great, but not an option from my E140 starting point.

I've been looking at this thread over on LH, and I'm not sure there's much advantage to leaving off the mass ring and making a 2234H instead of a 2235H. You'd lose a few db from the low end, get a bit better mid-band efficiency, and wind up with a very similar curve around the crossover. Unless the marginally better efficiency is integral to your design, I can't see a reason to go without the mass ring. With it, you get better bass extension and a slightly flatter response curve. Am I missing something?

 

[billyunderhi]

Re-Drawn Z 19

Someone suggested redrawing the Z-19
Here is my rendition


Comments and input on improvements would be welcome

Check the wiring over and let me know if there are mistakes.

I am making an assumption on the Lines In from speaker posts.
I think it is correct.

Thanks

 

[Zonker92]

That's pretty cool, Billy! Thanks for taking this on!

If I can suggest some changes without being a dick, I would try to clean up the wire routing as much as possible, and use more straight lines than hand-drawn wiggly ones. Also, for C2, I think you can just show it as 8 uF. No need to break it into two separate caps I think. As you said, my hope is to rationalize and simplify the wiring to keep the runs as short and tight as possible, like this one:

What are you wondering, about the L pads? We discuss them some around here:

http://www.audiokarma.org/forums/showpost.php?p=4343186&postcount=160

http://www.audiokarma.org/forums/showpost.php?p=4343186&postcount=161

 

[GordonW]

That was a comment from Zilch when he was measuring my speaker. Said he didn't know if the 2235 would be happy up to 1200 and suggesting that I make a Z-19 and try in out.
Steve

It seems like the 2226 magnet assembly (vented gap) and the voice coil I used, had somewhat "better linearity" around crossover, with the 2235 cone, than the stock 2235 assembly. It should definitely go a LITTLE higher, flat, than the 2235.

Regards,
Gordon.

 

[billyunderhi]

Zonker

Looking at the diagram as is can you tell me from the terminal strip

Lpad 1 Post one to Terminal G
Lpad 1 Post two to Terminal 2

Lapd 2 Post one to Terminal ?

I will add the L Pads to the Drawing then.

I do plan on making a clean line drawing when I know for sure how this gets set up. Thanks-

 

[Zonker92]

Cool!

Lpad 1 Post one to Terminal "G"
Lpad 1 Post two to Terminal "4"

Lapd 2 Post two to Terminal "2"
Lapd 2 Post one to Terminal "1"
Lapd 2 Post three to Terminal "3"

See the top-right L-pad notations on this drawing:

BUT I see no reason on earth to retain the confusing "G, 4, 2, 1, and 3" designations. I suspect Zilch did that because it made sense to him for other reasons. I think it would be much clearer to discard those designations and simply to label the five points on your top right terminal block as shown in the notations on the margin of the drawing above, that is, M1, M2, H1, H2, and H3, then run those terminals to the corresponding posts on the Mid and High L-pads. That's how I marked my boards.

 

[billyunderhi]

Post 232 is upgraded to show new crossover layout.

Zonker -did you run your 19s - Valencias with standard crossover?
How does the Mod 19 sound in comparison?

I have the 902 drivers as well - standard valencia style crossover

 

[Zonker92]

Dude, that is awesome! I'll have to pick through it soon for accuracy, unless you've done so already. Freaking strong work! I'm really pumped about that.

I ran my 846Bs with the stock crossover for a month or two and then put in a Z19 into one of them. Played A versus B there was a noticeable difference, but of course they both have L Pads (or something like an L Pad on the stock crossover), so it's kind of hard to compare without trying to get them equalized the same.

So anyway, it's really hard to discern or articulate accurate listening perceptions, but here's my subjective impressions, probably colored by everything I'd read here by Zilch and others: The stockers were a bit shrill and harsh in the midrange, and the highs rolled off. The soundstage was very forward. The Z19s seem more balanced, with better highs and more clarity and detail; the soundstage is farther back. Overall, definitely a large, noticeable improvement to my ears.

I did not, however, notice more bass, and in my opinion the 846Bs are quite bass shy. I still use a sub to supplement the bass. Eventually I may go to Model 19 cabinets to get more bass.

But back to your drawing: WOW. :thumbs::thumbs::thumbs::thumbs:

 

[4343]

Cool! ...
BUT I see no reason on earth to retain the confusing "G, 4, 2, 1, and 3" designations. I suspect Zilch did that because it made sense to him for other reasons. I think it would be much clearer to discard those designations and simply to label the five points on your top right terminal block as shown in the notations on the margin of the drawing above, that is, M1, M2, H1, H2, and H3, then run those terminals to the corresponding posts on the Mid and High L-pads. That's how I marked my boards.

Those other reasons are that it tells you where to run the wire.

If you follow the lines from the numbers, there are matching numbers on the diagram. The 1, 2, 3, & 4 all correspond to points on the diagram. The only one that doesn't is "G", because I re-drew the line that used to go to ground... (It used to run across the top to the first terminal, the one that has a little "G" next to it.)

I agree that "H1", etc. is clearer. Even those though tell you which terminal on the L-Pad to connect to.

BTW, 4.7uF and 3.3uF do indeed add to 8uF, but 8uF is a lot harder to find than the very common values of 3.3uF and 4.7uF. So, again, it's that way for a very good reason. It's meant to be a picture of what to put where, not a schematic. Not that it does not have to follow the schematic, but in practice some parts are easier to find than others...

 

[Zonker92]

Thanks for the input, 4343; I stand corrected. Billy, do you have any interest in breaking the 8 uf back up into 3.3 and 4.7? And 4343, should we break the 21 uf back into 9.1 and 12, or are 21s very common?

Sorry about that.

 

[billyunderhi]

No prob - I think I will leave the schem drawing as is for now.

Please look it over - I can make all additions at one time.


I found Zilchs discussions on the z-19 model 19s and Valencias about two weeks before he passed.
At the time I had just bought a set of Heathkit 101s for my dad. They needed recap and all roads lead to Zilch when you google those Altec items.

I am back to basic/stock but new capacitors and I found zilch's Z-19 schematic on another site. I couldn't sort out the L pad stuff well enough and then - I found this link. The drawing gave insight but there were a few things missing that I needed. I don't have alot of crossover build exp -

There are great parts lists and discussions on what to buy and add together for this upgrade.

I will be using a 7.0 oil capacitor and a 1.2 WIMA MPK to make 8.
A 20 Mf oil and a 1.2 WIMA for the 21.

The Heathkits sound great with stock values - however the LPad s first quarter turn seems to be where any horn control is...not much change after that.

I think the dual Mid/High control will help fine tune.

Thanks for the compliment Zonker
I'll put a note in the margin about Capacitors and combination of values.

Cant wait to build now.

 

[Zonker92]

Nice! Please keep us updated; I love this stuff. I have made one set of Z19s with Daytons, but now I'm thinking of doing it again from your drawing (I'll go over it soon to see if I have any more suggestions) using fancier caps and a cleaner layout than what I did before.

I'm not sure how the Heathkit crossover differs from the Altec N800s, but I recall reading somewhere that the Altec crossover for Valencias (not sure if it was all of them or only certain models) did not employ a true L Pad. A true L Pad is supposed to maintain the same impedance at every position, while I recall reading that the Valencia crossover instead presented varying impedances at different settings. Not sure how that plays into your experience with the Heathkits.

Thanks again!!!!

 

[4343]

Thanks for the input, 4343; I stand corrected. Billy, do you have any interest in breaking the 8 uf back up into 3.3 and 4.7? And 4343, should we break the 21 uf back into 9.1 and 12, or are 21s very common?

Sorry about that.

No, no, no need for sorry! Learning is what this is all about, I'm just here to "point out the obvious". That it's not obvious to everyone is part of the learning curve. It wouldn't look obvious to me if I hadn't had it pointed out countless times by those who knew...

And yes, the 21uF is another case of an odd value (unobtainium probably) being made up from the common values you see on the parts list, but NOT the schematic. Thanks for pointing that out, I missed that... Goes to my point that that drawing was not intended to replace the schematic, it's a _parts placement_ drawing.