I would like to build the Finemet crossover for Altec A7 as described here: https://theaudiofeast.com/blog/the-...ttkcSTuWW5KlrrEYrbnddcHIlVzW8o2ZJY_5E2uEQRZG4
It uses an autoformer and its crossover frequency is 850hz.
My Altec A7 has the following components:
My understanding is the 803b horn is designed to go down to 300hz, as the 3 in 803 indicates.
It is very unclear to me yet how autoformers and crossover in general work.
Is there any change I do to the Finemet crossover design to lower the crossover point from 850hz to a lower value such as 300hz?
Attached schematic and picture taken from the original article on theaudiofeast.com.
Thanks!
It uses an autoformer and its crossover frequency is 850hz.
My Altec A7 has the following components:
- Altec 288c 8ohm compression driver
- Altec 803b horn
- Altec 421-8H woofer
My understanding is the 803b horn is designed to go down to 300hz, as the 3 in 803 indicates.
It is very unclear to me yet how autoformers and crossover in general work.
Is there any change I do to the Finemet crossover design to lower the crossover point from 850hz to a lower value such as 300hz?
Attached schematic and picture taken from the original article on theaudiofeast.com.
Thanks!
Mmhh, I've Just replied to a similar question
https://www.diyaudio.com/community/...ing-dynaudio-2-way-setup.396717/#post-7290138
This particolar One exhibits an autoformer and a multi-tap coil.
Besides that, I don't think the 3 in 803 means 300 Hz capability.
https://www.diyaudio.com/community/...ing-dynaudio-2-way-setup.396717/#post-7290138
This particolar One exhibits an autoformer and a multi-tap coil.
Besides that, I don't think the 3 in 803 means 300 Hz capability.
Hi, I can’t help with the autoformer but just a couple things to think about. None of the components you list are from an A7 system with the possible exception of what you haven’t listed, the cabinet. Hopefully it’s an 828 so it will have the appearance. Yes the 3 in the 803 horn means 300 Hz but the 288C won’t do 300. They list it as 500 but I have found it seems to enjoy more like 800. The 421 8H is okay but more of an instrument woofer rather than sound reinforcement. If you have them fine, but if you have a choice and you need the power handling greater than the original A7 woofers, try the 421 8LF.
Hopefully others can help with the guts.
Hopefully others can help with the guts.
Last edited:
If I understand correctly, this is an autoformer shunted with a "swamping" resistor on the primary side. The autoformer (or transformer, works the same) reduces the voltage to the HF driver by trading voltage for current. So a 6 dB attenuation would mean 1/2 the voltage, and 2x the current, as seen by the HF driver. That would be a 2:1 turns ratio.
The other effect of the autoformer (or transformer) is a multiplication of impedance. A 2:1 turns ratio reduces the voltage by two times, but the impedance ratio is the square of the turns ratio, so the 8 ohms of the HF driver is multiplied 4 times, or 32 ohms. This increase in impedance messes up the crossover frequency, so a resistor is placed in front of the autoformer ... a swamping resistor, which swamps out the increase in resistance caused by the autoformer. At high attenuation levels, the resistance the crossover sees is set by the resistor, not the autoformer+HF driver combination.
In practical terms, the overall effect is similar to an L-pad, but the HF driver does see a bit lower source impedance compared to an L-pad, so it will sound different.
The other effect of the autoformer (or transformer) is a multiplication of impedance. A 2:1 turns ratio reduces the voltage by two times, but the impedance ratio is the square of the turns ratio, so the 8 ohms of the HF driver is multiplied 4 times, or 32 ohms. This increase in impedance messes up the crossover frequency, so a resistor is placed in front of the autoformer ... a swamping resistor, which swamps out the increase in resistance caused by the autoformer. At high attenuation levels, the resistance the crossover sees is set by the resistor, not the autoformer+HF driver combination.
In practical terms, the overall effect is similar to an L-pad, but the HF driver does see a bit lower source impedance compared to an L-pad, so it will sound different.
Hey Fluxus. What Cal says is correct, you don't have an A7, you have an A5. Why? Because the A7 used the 1" horns like the 811 or 511 and a 1" driver like the 802 or 806. The A5 is bigger on top, using the multicell horns like you have. I'd had the A7 with both those horns and both those drivers, I converted it to an A5.
Crossing the 288/803 at 300Hz? That's a stretch, to put it politely. On the 803 I have mounted the 288, 291, 290, early Radian and BMS coaxials. Even with the 290 "Giant Voice" and the BMS coaxials which are specified down to 300Hz, it didn't really go that low, at least not cleanly. The horn should, but I could never get the drivers to do it well. IME you'll be much happier with a crossover point an octave higher, circa 600. Hz. I do understand the desire for the low crossover point on the horn, but if it doesn't work well, then there is no reason to do it. If it sounds better crossed at 500 or 650, then do that.
Crossing the 288/803 at 300Hz? That's a stretch, to put it politely. On the 803 I have mounted the 288, 291, 290, early Radian and BMS coaxials. Even with the 290 "Giant Voice" and the BMS coaxials which are specified down to 300Hz, it didn't really go that low, at least not cleanly. The horn should, but I could never get the drivers to do it well. IME you'll be much happier with a crossover point an octave higher, circa 600. Hz. I do understand the desire for the low crossover point on the horn, but if it doesn't work well, then there is no reason to do it. If it sounds better crossed at 500 or 650, then do that.
@tubularfluxus Do you currently have a crossover that you are using? If so, what is it?
The crossover you posted above is a first order at approximately 850 Hz. A first order crossover is not normally used in a rig like this, more often it's second, third or forth order crossover. Especially if you cross low, you'll want a steeper filter to keep the low frequencies out of the driver. In a passive crossover I often use 3rd order electrical, which results in 4th order acoustic, more or less.
The best thing to do is to measure. Do you have measurement capability? Or would you like to acquire it?
The nominal 2nd order crossover at 600Hz and 8 ohms is 16.2uF capacitor in series and 4.3mH inductor to ground. With that you'd use either a transformer or L-Pad to attenuate the 288 down to the level of the woofer.
Altec named their multicell horns with the number of cells first, then a zero, then the cutoff frequency. For example, the 1505 is a 500Hz horn with 15 cells, the 1005 is 500Hz with 10 cells, the 203 a 300Hz horn with 2 cells. There were 300, 400 and 500Hz horns. I've owned and used a few of the 300 and 500 Hz horns, and have never seen a 400Hz multicell in person.
The crossover you posted above is a first order at approximately 850 Hz. A first order crossover is not normally used in a rig like this, more often it's second, third or forth order crossover. Especially if you cross low, you'll want a steeper filter to keep the low frequencies out of the driver. In a passive crossover I often use 3rd order electrical, which results in 4th order acoustic, more or less.
The best thing to do is to measure. Do you have measurement capability? Or would you like to acquire it?
The nominal 2nd order crossover at 600Hz and 8 ohms is 16.2uF capacitor in series and 4.3mH inductor to ground. With that you'd use either a transformer or L-Pad to attenuate the 288 down to the level of the woofer.
It does mean 300Hz (as Cal points out). Cal owns a pair of the giant 1803 horns, which are also 300Hz.
Altec named their multicell horns with the number of cells first, then a zero, then the cutoff frequency. For example, the 1505 is a 500Hz horn with 15 cells, the 1005 is 500Hz with 10 cells, the 203 a 300Hz horn with 2 cells. There were 300, 400 and 500Hz horns. I've owned and used a few of the 300 and 500 Hz horns, and have never seen a 400Hz multicell in person.
The Altec 290 "Giant Voice" driver with the phenolic diaphragm is rated to play down to 300 Hz on the big 300 Hz horns. I could not get a good response down there. Maybe OK for a baseball stadium announcements, but not for Hi-Fi. It worked well crossed nearer 450 Hz. I'll see if I can dig up the measurements.
More important than the autoformer (at the moment) is the actual high pass filter and EQ for the 288 on the 803. A simple first order high pass isn't a good idea. There is more to it than that.
More important than the autoformer (at the moment) is the actual high pass filter and EQ for the 288 on the 803. A simple first order high pass isn't a good idea. There is more to it than that.
Yes indeed, the impedance of the driver will appear to be higher and higher as the attenuation is increased. Looking at the values given in the table for R1, they seem pretty good by my calculations. I even took the measured impedance curve of a 16 ohm 288 on an 803 horn and magnified it by 15.84X (the -12dB point). A 17.8R swamping resistor does a good job of presenting a fairly flat impedance to the amp of 16-17.5 ohms.
Lol yes it was an Altec A7, but I have been replacing few parts:
I am replacing the Altec 802-8T I have for Altec 288c 8ohm with GPA diaphragms
I am replacing the Altec 421-8H I have for either Altec 416 or Altec 515. Haven't made this purchase yet.
I am replacing the Altec 511 horns I have for Altec 803b, haven't made this purchase yet.
Sounds like I should just keep the crossover point to 850hz, like this crossover design does.
I am replacing the Altec 802-8T I have for Altec 288c 8ohm with GPA diaphragms
I am replacing the Altec 421-8H I have for either Altec 416 or Altec 515. Haven't made this purchase yet.
I am replacing the Altec 511 horns I have for Altec 803b, haven't made this purchase yet.
Sounds like I should just keep the crossover point to 850hz, like this crossover design does.
By the way, I've also decided to add a super tweeter, either a JBL 077 I have, or a Pioneer PT-R9 I would like to source.
Any tip of how I could add a super tweeter?
I have seen on similar crossover design this done by connecting the supertweeter positive to one of the tap of the autoformer, passing through a 1.5uf cap.
Any tip of how I could add a super tweeter?
I have seen on similar crossover design this done by connecting the supertweeter positive to one of the tap of the autoformer, passing through a 1.5uf cap.
You don't need to keep the 850 crossover, but as it's first order it's worth a try. I think you can do better.
Yes you can usually just had a cap in series with super tweeter and tap off the 288. Maybe not the most elegant plan, but it should work OK
Yes you can usually just had a cap in series with super tweeter and tap off the 288. Maybe not the most elegant plan, but it should work OK
yep swamped "primary" autoformer - maybe some subtle harmonics added by the autoformer to sweeten the horn. Take away the 13.6 ohm swamping resistor and a smaller , perhaps more "exotic" cap can be used (or charge coupling with two series cap and an applied DC via resistor at their junction) - I bet some engineer would suggest resistor padding.
Except this AF application and the One on some models of RogerS LS 3/5a, the use of autoformer I've seen Is mostly on woofers, not tweteers.
This Is because you want plenty of bass but not too much, especially if you put a, say, 10" woofer near the floor and place the 5"mid and 1" or 3/4" tw at ear eight.
This Is because you want plenty of bass but not too much, especially if you put a, say, 10" woofer near the floor and place the 5"mid and 1" or 3/4" tw at ear eight.
Hmm, the 511/802 acoustic pathlength is ~18.875+4.56+1.5 = 19.94"+ end correction = 2*(7+18.5) = 25.5/pi/2 = ~8.12*0.613 = ~5"/4pi space, so a bit more for any baffle surfaces.
19.94+5 = ~25" or ~13543/4/25 = ~135 Hz cutoff and since we need at least -24 dB for driver protection a 2nd order requires it to be ~135*2^2 = ~542 Hz, so normally 500 Hz is fine once on a baffle, which originally it was until the 'bean counters' forced compromises.
Anyway, from this a 1st order requires ~135*2^4 = 2160 Hz, so at 850 Hz they're in theory relying a lot on ~flea power/impedance matching/AF shelving to keep it from self destructing.
The 803/288 goes down to around 75-80 Hz for a 300 Hz/2nd XO, ergo 1200 Hz/1st, but its shelving alone should be enough to safely allow an 850 Hz XO or lower if coupled to a matching impedance.
19.94+5 = ~25" or ~13543/4/25 = ~135 Hz cutoff and since we need at least -24 dB for driver protection a 2nd order requires it to be ~135*2^2 = ~542 Hz, so normally 500 Hz is fine once on a baffle, which originally it was until the 'bean counters' forced compromises.
Anyway, from this a 1st order requires ~135*2^4 = 2160 Hz, so at 850 Hz they're in theory relying a lot on ~flea power/impedance matching/AF shelving to keep it from self destructing.
The 803/288 goes down to around 75-80 Hz for a 300 Hz/2nd XO, ergo 1200 Hz/1st, but its shelving alone should be enough to safely allow an 850 Hz XO or lower if coupled to a matching impedance.