Hiraga 600hz (A5) Crossover - Sound Practices

Since the crossover is designed for an 8ohm driver, you will want to use a 'swamping' resistor across the primaries of your autoformer that 'present' an ~8ohm impedance to your crossover.

Based on intact audio and the calculator there, for an 8ohm driver, 8ohm crossover impedance, and ~12dB of attenuation you would need an ~8.5ohm 'swamping' resistor across the primaries of the autoformer.

I think that's right, and hope others will chime to confirm. FWIW, it's the plan I'm working on implementing on my crossovers in the next week or so. The difference for me is I'm using Werner Jagusch's (sp?) autoformers.
Yes, that´s right. From -8dB on up there is very ltttle difference in the required value of the swamping resistor. I think 9 Ohms would be a good average size from -8dB upwards.
 

marantzfan

Administrator
Staff member
Started making a layout as a I usually do be before building a crossover. I don't know why, but for some reason I am struggling with this one. On paper it certainly doesn't look that complicated, but it quickly becomes a spaghetti mess when I add all the wire connections. Ugh.

This is what I have so far, but I know it's overly complicated and I'm not sure how to incorporate the auto former.

Anyone smarter than me, feel free to double check my work. ;)

1593379222883.png
 
Last edited:

marantzfan

Administrator
Staff member
Yes, that´s right. From -8dB on up there is very ltttle difference in the required value of the swamping resistor. I think 9 Ohms would be a good average size from -8dB upwards.
The other resistors in the circuit are all mills; so I'll get a pair of 9 ohm mills ordered for the swamping resistors as well.
 

gable

Senior Member
Started making a layout as a I usually do be before building a crossover. I don't know why, but for some reason I am struggling with this one. On paper it certainly doesn't look that complicated, but it quickly becomes a spaghetti mess when I add all the wire connections. Ugh.

This is what I have so far, but I know it's overly complicated and I'm not sure how to incorporate the auto former.

Anyone smarter than me, feel free to double check my work. ;)

View attachment 24960
I think that's a decent layout really. The autoformer looks to be in a good spot, as it's the last piece before the outputs.
 
The other resistors in the circuit are all mills; so I'll get a pair of 9 ohm mills ordered for the swamping resistors as well.
Are you sure that you will need the notch filter when you connect the 3636 and the swamping resistor? Will you be able to do measurements?
Pehaps listening with or without the notch filter would be a good idea. To simplify using the 3636 I think 2 three way switches will be sufficient, as you are unlikely to need the 0 - 4 (1 to 3 dB) setting on the output.
 

marantzfan

Administrator
Staff member
Are you sure that you will need the notch filter when you connect the 3636 and the swamping resistor? Will you be able to do measurements?
Pehaps listening with or without the notch filter would be a good idea. To simplify using the 3636 I think 2 three way switches will be sufficient, as you are unlikely to need the 0 - 4 (1 to 3 dB) setting on the output.
That would certainly simplify things. So essentially, eliminating all of this? Curious, how does the addition of the auto former replace the need for a notch filter?

1593447931222.png


Btw, I read through that thread that you mentioned by Panomaniac on Hostboard. I skimmed through all 25 pages but never saw anything approaching a published schematic.
 
That would certainly simplify things. So essentially, eliminating all of this? Curious, how does the addition of the auto former replace the need for a notch filter?

View attachment 24987


Btw, I read through that thread that you mentioned by Panomaniac on Hostboard. I skimmed through all 25 pages but never saw anything approaching a published schematic.
'
 
No guarantees, but the autoformer + swamping resistor gives a very solid 8 Ohm impedance looking into the driver and looking back into the crossover. Worth a try. The crossover in Panos thread that I referred to is on page 13, post #127.
 

gable

Senior Member
No guarantees, but the autoformer + swamping resistor gives a very solid 8 Ohm impedance looking into the driver and looking back into the crossover. Worth a try. The crossover in Panos thread that I referred to is on page 13, post #127.
I think it's a good idea to try removing the notch filter when you have the autoformer and swamping resistor in place. The swamping resistor provides a more stable impedance to the crossover, and may mitigate the need for the notch filter, as @Boli46 mentioned.
 

marantzfan

Administrator
Staff member
I think it's a good idea to try removing the notch filter when you have the autoformer and swamping resistor in place. The swamping resistor provides a more stable impedance to the crossover, and may mitigate the need for the notch filter, as @Boli46 mentioned.
I've seen where even Pete Riggle has adopted another version of this crossover where he added a three position switch which allows the user to either affect the full -4db, -2db or remove the notch filter entirely. Interestingly in the response curves on his site, the curve is the flattest without the notch filter...


Certainly food for thought and he is using an L-Pad. not an autoformer.
 

billfort

Administrator
Staff member
I'm getting lost here. I thought one of the main 'features' of the Hiraga style crossover was to leverage the high-efficiency of that big horn and use controlled attenuation – with gradually increasing high frequency bypass of that attenuation – to flatten overall response while extending apparent high frequency response.

From the 'Taming the A5 Classic for domestic use' article;

"L3, and C3 along with R3, R4 and R5 make up a resonant trap with controlled Q and attenuation to equalize the 112 dB midband response of the 288 driver down to 100 dB and to allow it to go to 16 kHz with little attenuation."​

I'm not seeing how eliminating that 1.7uF cap that bypasses a good portion of the attenuation would still give you that rising response.
 

marantzfan

Administrator
Staff member
I'm getting lost here. I thought one of the main 'features' of the Hiraga style crossover was to leverage the high-efficiency of that big horn and use controlled attenuation – with gradually increasing high frequency bypass of that attenuation – to flatten overall response while extending apparent high frequency response.

From the 'Taming the A5 Classic for domestic use' article;

"L3, and C3 along with R3, R4 and R5 make up a resonant trap with controlled Q and attenuation to equalize the 112 dB midband response of the 288 driver down to 100 dB and to allow it to go to 16 kHz with little attenuation."​

I'm not seeing how eliminating that 1.7uF cap that bypasses a good portion of the attenuation would still give you that rising response.
I read a description on another forum from @J-ROB where he describes what all is happening with this notch filter, and it was pretty interesting. I'll see if I can find it and link it here.
 

marantzfan

Administrator
Staff member
Found it. Keep in mind, he is referring to the 16 ohm version. so component values differ:

The EQ in the crossover has two features. L3 + R4 is a broad "notch" to shelve the 3k peak and the 24 ohm + 0.82uF is a classic peaking circuit.

The peak happens because as the frequency rises, the cap comes into play and bypasses the 24 ohm resistor, which forms a voltage divider with the 10 ohm. As frequency increases it is "turning down" the 24 ohm like a volume control, so to speak.

At low frequencies it is like the cap is not there and you only have a 2 resistor voltage divider, At high frequencies the 24 ohm "dissappears" due to the cap shorting it out for AC. At 8kc, the 0.82uF cap offers 24 ohms impedance so half of the resistor is "disappeared."

If you slightly increase the 24 ohm to 28 or 30 ohms, you will have more attenuation at the lower frequencies where it is a pure 2 resistor voltage divider and the same output at the top end because R3 is "disappeared."

I did exactly this with my Hiraga crossover (on 288C and 1505) but it was many moons ago and I forget the details. I found the lower mids to be slightly too hot with the stock values, maybe in part because I was using an Onken/416 bass cab not a VOT box.
 

billfort

Administrator
Staff member
Still confused. J-Rob said; "...and the 24 ohm + 0.82uF is a classic peaking circuit".

If you take out the 1.7uF cap circled in red, what cap is doing the high-frequency bypassing of the attenuation?
 

Audionut

Next Round Is On Me
🍿, can’t wait to see the autoformers implemented. I have my reading cut out at intact audio to properly implement mine.
 

marantzfan

Administrator
Staff member
No guarantees, but the autoformer + swamping resistor gives a very solid 8 Ohm impedance looking into the driver and looking back into the crossover. Worth a try. The crossover in Panos thread that I referred to is on page 13, post #127.
That link takes me to a French audio forum, which I couldn’t make out...

I found this though, look at post #19:

 
That link takes me to a French audio forum, which I couldn’t make out...

I found this though, look at post #19:

The link took you to the right place. Just scroll down the first post and click on the two stamp sized circuits to enlarge them.
 

marantzfan

Administrator
Staff member
The link took you to the right place. Just scroll down the first post and click on the two stamp sized circuits to enlarge them.
Ok, so I've found those. They both appear to be 3-way filters and the one with values listed appears to be for TAD mids and JBL tweeters. I'm not sure how I would adapt this circuit to a 2-way design and the drivers I am using.

Attaching both schematics here for the sake of group discussion, also here is a link to the thread being referenced:



1593530525545.png

1593530375356.png
 
Last edited:
Ok, so I've found those. They both appear to be 3-way filters and the one with values listed appears to be for TAD mids and JBL tweeters. I'm not sure how I would adapt this circuit to a 2-way design and the drivers I am using.
They both show the same filter. You can’t make it into a two-way, but you are not limited to the drivers in the schematic. It is supposed to be pretty much textbook for the crossover points used.
 

marantzfan

Administrator
Staff member
They both show the same filter. You can’t make it into a two-way, but you are not limited to the drivers in the schematic. It is supposed to be pretty much textbook for the crossover points used.
Got it. So, really the only way to consider this as an option is to add a tweeter. Unfortunately that is not something I'm considering right now.

So, I guess that leaves us back at the original circuit being discussed for a 2-way solution.
 
Top