Yet Another Altec 2-Way

[mdye]

Could I also remove the 13R resistor and move the 16 ohm L-pad over in its place (only using terminal 1 and 2) as discussed here: Altec Attenuation Circuit Help

Would this give me the option to adjust the midrange to taste?

 

[EarlK]

@mdye

You can, but the adjustments end up being pretty wonky/sensitive ( to the point of futility > IMHO ).

I'll show you later today what I mean.

Best for now is to take out the 13R resistor ( both chnls ) and do some listening.



PS; Here's the predicted difference between keeping in place the 13R resistor ( vs removing it which is dark green in the pic ).

- The 16 ohm Variable LPad is running wide open ( no attenuation ).

BTW, your 806a likely doesn't have this much UHF ( most take a dive after 8K ).
- This is an example of an outlier that must have had a "lite" version of the 20275 in it
- ( I've only ever seen 1 other example that measured this well ).

 

[mdye]

The crossover for the bi flex starts the slope early at around 800 hz.

Right, which I figured would mate well with the HF section of the N-800-F, or am I completely mistaken?

 

[mdye]

So, I disconnected one end of the 13R resistor, and the midrange definitely came back, but it came back with a vengeance. Crossover beginners question: can you safely swap components in and out of a crossover while playing music? So, for example, could I A/B test by connecting and disconnecting the 13R resistor in and out of circuit without doing damage to a costly and original 806 diaphragm?

 

[EarlK]

So, I disconnected one end of the 13R resistor, and the midrange definitely came back, but it came back with a vengeance. Crossover beginners question: can you safely swap components in and out of a crossover while playing music? So, for example, could I A/B test by connecting and disconnecting the 13R resistor in and out of circuit without doing damage to a costly and original 806 diaphragm?

Don't

Now I'd do the live-swapping ( if it was me doing it ) >>> but truthfully ( based on your noob-ness ), I'm not confident that you won't simply touch the wrong thing, create a short-circuit and then smoke something >> so, better safe than sorry.

Questions like that are a lot like " is it safe for me to poke around inside a freshly used tube amp " ( only enter if you're very experienced ).


You're best bet here is to replace the 13R resistor with something larger ( between 20R - 30R ) and then wire the variable LPad in series with said resistor, then wire terminal 1 ( of the Lpad ) to ground.
- Use terminals 2 + 1 on that 16 ohm Lpad.

That'll give you a variable resistance range of something like 20R up to 80R .

Play with midrange levels with this setup till your happy with the mid balance.

Remove the Lpad and measure the resistance between 2 + 1 ( if you want to replace with a fixed resistor in that position ) .

 

[mdye]

Really appreciate the detailed assistance here, Earl! I'll give this new arrangement with the Lpad a shot this weekend and report back.

 

[Phillip Jeffries]

I’ve got the 419-8b bi flex in a variation of the je labs 2 way 32b horn with 802-8g.

did a few variations of crossover. The series crossover works really well. I crossed over in the 2.8-2.9k range. Basically, there was a natural dip just below that in the bi flex, so that seemed a good spot. These drivers are 8 ohm, so I think the crossover is .27 mh with 8-10 mf, can’t remember exactly. Then the l pad to the horn with a 2 mf across the series section for some high frequency boost.

 

[mdye]

I’ve got the 419-8b bi flex in a variation of the je labs 2 way 32b horn with 802-8g.

did a few variations of crossover. The series crossover works really well. I crossed over in the 2.8-2.9k range. Basically, there was a natural dip just below that in the bi flex, so that seemed a good spot. These drivers are 8 ohm, so I think the crossover is .27 mh with 8-10 mf, can’t remember exactly. Then the l pad to the horn with a 2 mf across the series section for some high frequency boost.

Interesting. Was the basis of this the same series crossover from post #2 in this thread?

 

[Phillip Jeffries]

Yes, definitely the JE Labs info was the starting point and main inspiration. There was some other info I found on the web about the series crossover, too.

here are two screenshots from the analyzer app. The first, I think...I’m going back years in my camera roll, is the 419-8b alone, the second is one I took a few months ago of the speaker with crossover and horn.

 

[mdye]

@EarlK - does the exact placement of the 20R resistor in series with the Lpad matter? For example, could I wire terminal 2 of the Lpad to the point where the two 8R resistors connect and then wire the 20R resistor from terminal 1 to ground? Reason I ask is because it would make for a much simpler re-wire based on how I currently have things soldered up.

 

[EarlK]

@EarlK - does the exact placement of the 20R resistor in series with the Lpad matter? For example, could I wire terminal 2 of the Lpad to the point where the two 8R resistors connect and then wire the 20R resistor from terminal 1 to ground? Reason I ask is because it would make for a much simpler re-wire based on how I currently have things soldered up.

My picture shows the 20R resistor wired to the junction of the two 8R resistors.

It doesn't matter if the free end of that 20R resistor is connected to terminal 2 ( or terminal 1 ) of the variable Lpad.
- Obviously the previous choice made determines which other terminal ( of the LPad ) connects to ground .

The "best" reason to choose one terminal over the other is so that when that dial is cranked counter-clockwise the measured resistance ( between 2 + 1 ) goes down ( which will pull down the midrange ).

IOW, turning the LPad dial to the left matchs convention and expectation ( and midrange level is lowered ) .

Apart from that "convention" it doesn't matter at all what order those 2 resistive parts are connected together ( as long as one end of the R-string connects to the junction of the twin 8R resistors and the other end connects to ground .

 

[mdye]

That makes perfect sense. Thanks, Earl!

 

[EarlK]

How's this developing ?

Do you have an update to offer us ?

 

[mdye]

I sure do, Earl!

I made the Lpad move that we've been discussing, and I'm liking the range and ability to adjust the midrange to taste. I think that was a good move for my situation. That said, I'm still tweaking the HF a bit, which seemed a little too hot with the LPad removed as a sort of "global" Lpad on the 806A. Went back to studying other, similar crossover schematics and noticed that most of them had a 2-4R resistor in series between the 30923 network and the compression driver. I'm guessing this tames HF a bit. Decided to experiment, but the lowest value I could hobble together via parallel resistors was 4.5R. I gave that a try, but it sounds like a little too much padding. Therefore, I've ordered a bunch of 1R 10W ceramic resistors in order to dial in something that sounds well balanced. Those resistors should be here Wednesday, and I feel like I'm real close to having a network that is deserving of some serious, long term listening before I take any other actions.

One thing I should add is that I'm also experimenting by adding more capacitance in the LF crossover (in my schematic was 2mH/10uf) to better blend the woofer and compression driver. I'm currently trying 20uf, and I'm liking what I'm hearing so far.

👍

 

[mdye]

Just another follow up on this project. After some experimenting with varying resistances in series with the 806a, I've settled on 2 ohms which sounds very well balanced to me. I think I now have a crossover network that I can listen to for a while. I'll post an updated schematic as soon as I can, but for now I can share this RTA taking off my iPhone. This is with the aforementioned network, Altec 412Cs, 806As and 811B horns. About 1 meter away with pink noise. I see some steep dips in a few spots through the spectrum, and keeping in mind that my iPhone and living room acoustics aren't the last word in controlled measurements, I am left wondering: what is the acceptable +/- range for such a measurement of a speaker to be considered "flat?"

 

[Katalyst]

Nice graph in my opinion. Is this with the woofer at the bottom of the cabinet and horn on top?

Attachments​

• 

 

[mdye]

Nice graph in my opinion. Is this with the woofer at the bottom of the cabinet and horn on top?

Attachments​

• 

I haven't had a chance to try playing around with flipping the cabinets yet, so this graph comes from the woofer at the bottom of the cab. A quick picture of the test set up is attached.

 

[djnagle]

That graph looks pretty darn good. Depending on how old you are, the high end looks very extended for that driver. that dip-hump in the 50 to 70 hz range might flatten a bit if you flip your cabs over. With the driver low in the cab you might be getting some floor interference.

Measurement aside, how does it sound?

 

[mdye]

Sounds is pretty darn good so far. I finally got around to flipping the cabinets (placed them on some floor dollies), and that definitely helped to further blend the the horn with the 412c. I also picked up a pair of RCF H3709 horns for cheap, and I hate to say it, but I do prefer their sound over the 811s. I'm probably heading down a rabbit hole that will eventually lead to 32s.

One thing I've been doing is flipping back and forth between the crossover I built and a simple 1.5uf cap to the 806a. Both approaches seem to have their own merits sound-wise, but I think I prefer the crossover a bit more in this particular configuration. One thing I'm considering is building a second pair of crossovers using better quality parts--mainly using PIO caps instead of the inexpensive polys that are currently in there. I've got a lot of experience building amps and such with classic PIO caps like Vitamin Qs and old Russian stock, but I've never used motor run caps before. Are there certain brands that are preferred for one reason or another (reliability, etc.) or are they all basically the same when used in a crossover? Looking at various sites that sell such caps, and prices seem to be all over the place--you can get a 10uf motor run cap for $7 or $40, but I have no idea if the more expensive ones are better or not for audio applications.

 

[EarlK]

@mdye

Thanks for your recent updates.

It's very gratifying to see sonic success from someone using the 806 driver .
- Remind me please, are your 806's still loaded with 20275 diaphragms ?

I know what you mean ( I think ) about the sound from that sort of horn profile ( H3709 , 32A-C ).
I have some ( super rare ) NorthWest Model 340 horns that share the basic bell profile .
- The 340's are a straight 2" entry ( made for a TAD 4001 or JBL 2441 drivers back in the day ) and have 3 small vanes in the throat.
- They are extremely open sounding when compared to other large horns ( these are around 22-24" wide ).

This horn design saw limited use with only a couple of sound companies ( for example; back in the 1980-90's Maryland Sound and NWS ) and were also used for a limited time in at least one generation of George Augspurgers studio monitors.

Here's the only modern studio install that I know about that uses the Northwest 340 horn.