I have a 1971 Fisher S-695-IP console, with the following speaker compliment per side:
Qty 1 - 12” Rola woofer, 8 Ohm
Qty 2 – 5” Rola mid-range speakers, 8 Ohm each, wired in parallel (effective 4 Ohm)
Qty 1 – 3” CTS tweeter, 16 Ohm
The crossover is comprised of the following:
Woofer: 2.6 mH air coil/inductor
Mid-range: 5.6 Ohm resistor, 25 uF capacitor, .5 mH air coil/inductor
Tweeter: 4.7 Ohm resistor, 4 uF capacitor
Since the console is now 50 years old (wow how time flies!
) the old capacitors in the crossover are due to be replaced. And of course normally that would be all that is required, especially if no changes in crossover design or driver are to occur.
And normally I would approach this simple rehabilitation with the intent to make no changes and keep the design and drivers as manufactured by Fisher in 1971... effectively just replacing the capacitors with the same values and move on.
However, I have a few observations that lead me to consider modifying the crossover and replacing the tweeter with something more modern.
Fisher speakers are known to be voiced, as many manufacturers do, for a specific style and taste. Fisher was known to voice their speakers more toward vocal, classical, and jazz versus many of the styles that are popular today. With that in mind it seems from my listening and other subjective reviews that Fisher speakers seem to slightly mute or roll off the upper mids and highs a bit to perhaps “soften” the sound. And this does seem to be the case subjectively when listening to this 1971 Statesman console.
In addition, the use of 3” paper cone tweeter likely plays a part, even after replacing the capacitor, and noting that prototyping its replacement with a small soft dome really pleasantly brightens up the sound.
So, my first thought was to replace the OEM CTS 16 Ohm paper cone tweeter with a soft dome. But finding an appropriate new 16 ohm tweeter is challenging, and I found none except for a very expensive horn that isn’t a good fit in several ways.
But that’s not a big deal theoretically of course, as I simply need to modify the capacitor for the tweeter to accommodate an 8 Ohm modern tweeter, and leave the rest of the crossover alone.
However, this is where I am asking for some expert analysis of the OEM crossover before I change anything.
If I run the OEM crossover points through some online crossover calculators, it seems that the factory advertised crossover points of 2500 & 400 Hz may not align with the capacitor and coil/inductor values that are actually installed. There seems to be a big frequency gap between the mid-ranges and tweeter, and it takes choosing some wildly different crossover values to make the OEM capacitor & coil/inductor values to get even close... but not at the same time.
But I’m admittedly not an expert crossover designer, and along with the capacitors & coils involved, there are also a couple resistors in play.
I know that typically a resistor in a crossover may be used to attenuate a driver, and normally isn’t used to change the impedance of a driver, but I do wonder if perhaps it has a slight effect on impedance and thus influenced these crossover capacitor & coil/inductor values in a way I don’t understand and the online calculators can’t take into account.
Also, I realize that we don’t know a lot about the drivers themselves other than the impedance, and obviously the manufacturer did have that data. So that leaves us at something of a disadvantage, but really in this case I’m only concerned about the mid-range and tweeter portions so the bulk of the T/S parameters probably aren’t that relevant.
But sensitivity is relevant, however without objectively testing the drivers or having their specifications we’ll likely never know, but I can say it is likely that they all are high sensitivity given their vintage and the relatively low power (30ish wpc) of the receiver.
Whew… OK 🙄
So can someone with some crossover expertise look at the OEM design and try to explain how it is working to deliver the specified 2500/400 crossover points, or discern if there are any issues with that design such that is is not delivering the specified (or more ideal) crossover points and if so what component values should be modified?
And with that in mind, and my plan to replace the 16 Ohm tweeter with an 8 Ohm tweeter, what component values would be appropriate to change in the crossover?
Thanks in advance!
Qty 1 - 12” Rola woofer, 8 Ohm
Qty 2 – 5” Rola mid-range speakers, 8 Ohm each, wired in parallel (effective 4 Ohm)
Qty 1 – 3” CTS tweeter, 16 Ohm
The crossover is comprised of the following:
Woofer: 2.6 mH air coil/inductor
Mid-range: 5.6 Ohm resistor, 25 uF capacitor, .5 mH air coil/inductor
Tweeter: 4.7 Ohm resistor, 4 uF capacitor
Since the console is now 50 years old (wow how time flies!
) the old capacitors in the crossover are due to be replaced. And of course normally that would be all that is required, especially if no changes in crossover design or driver are to occur.And normally I would approach this simple rehabilitation with the intent to make no changes and keep the design and drivers as manufactured by Fisher in 1971... effectively just replacing the capacitors with the same values and move on.
However, I have a few observations that lead me to consider modifying the crossover and replacing the tweeter with something more modern.
Fisher speakers are known to be voiced, as many manufacturers do, for a specific style and taste. Fisher was known to voice their speakers more toward vocal, classical, and jazz versus many of the styles that are popular today. With that in mind it seems from my listening and other subjective reviews that Fisher speakers seem to slightly mute or roll off the upper mids and highs a bit to perhaps “soften” the sound. And this does seem to be the case subjectively when listening to this 1971 Statesman console.
In addition, the use of 3” paper cone tweeter likely plays a part, even after replacing the capacitor, and noting that prototyping its replacement with a small soft dome really pleasantly brightens up the sound.
So, my first thought was to replace the OEM CTS 16 Ohm paper cone tweeter with a soft dome. But finding an appropriate new 16 ohm tweeter is challenging, and I found none except for a very expensive horn that isn’t a good fit in several ways.
But that’s not a big deal theoretically of course, as I simply need to modify the capacitor for the tweeter to accommodate an 8 Ohm modern tweeter, and leave the rest of the crossover alone.
However, this is where I am asking for some expert analysis of the OEM crossover before I change anything.
If I run the OEM crossover points through some online crossover calculators, it seems that the factory advertised crossover points of 2500 & 400 Hz may not align with the capacitor and coil/inductor values that are actually installed. There seems to be a big frequency gap between the mid-ranges and tweeter, and it takes choosing some wildly different crossover values to make the OEM capacitor & coil/inductor values to get even close... but not at the same time.
But I’m admittedly not an expert crossover designer, and along with the capacitors & coils involved, there are also a couple resistors in play.
I know that typically a resistor in a crossover may be used to attenuate a driver, and normally isn’t used to change the impedance of a driver, but I do wonder if perhaps it has a slight effect on impedance and thus influenced these crossover capacitor & coil/inductor values in a way I don’t understand and the online calculators can’t take into account.
Also, I realize that we don’t know a lot about the drivers themselves other than the impedance, and obviously the manufacturer did have that data. So that leaves us at something of a disadvantage, but really in this case I’m only concerned about the mid-range and tweeter portions so the bulk of the T/S parameters probably aren’t that relevant.
But sensitivity is relevant, however without objectively testing the drivers or having their specifications we’ll likely never know, but I can say it is likely that they all are high sensitivity given their vintage and the relatively low power (30ish wpc) of the receiver.
Whew… OK 🙄
So can someone with some crossover expertise look at the OEM design and try to explain how it is working to deliver the specified 2500/400 crossover points, or discern if there are any issues with that design such that is is not delivering the specified (or more ideal) crossover points and if so what component values should be modified?
And with that in mind, and my plan to replace the 16 Ohm tweeter with an 8 Ohm tweeter, what component values would be appropriate to change in the crossover?
Thanks in advance!
Attachments
Last edited:
Are you adding the series resistors' values to the drivers' impedances when calculating the crossover points?
Also the impedance is approximate, and will usually vary from the spec near the crossover points.
A new tweeter would require both new resistor and capacitor values that depend on the exact one chosen.
Also the impedance is approximate, and will usually vary from the spec near the crossover points.
A new tweeter would require both new resistor and capacitor values that depend on the exact one chosen.
Last edited:
I have run the OEM component values through a calculator like the one below, and played with adding all or a portion of the series resistors' values, but the OEM specified crossover points still seem to not align with the values I'm seeing. And when I can get the tweeter or mid-range value to match the OEM crossover point, the other (tweeter or mid-range) is nowhere close, which would seem to create a big frequency gap.
3-Way Crossover Calculator / Designer
I do understand how driver impedance values vary through the frequency range, and that the new tweeter would require a new capacitor and possibly resistor. Based on the existing crossover design where a 4 uF capacitor is used with the 16 Ohm tweeter, an 8 uF capacitor should be the right fit, again though if the rest of the circuit is really built correctly from Fisher.
If the resistors weren't already in the mix from OEM, and if the capacitor and coil/inductor values didn't seem to be possibly wrong, it would be easy to model what I need. And if either the resistors weren't there OR the possibly wrong capacitor and coil/inductor values seemed correct, this would be easy for me figure out. However both in combination are causing me to wonder about the design and make me uncertain about the best replacement capacitor and possibly coil/inductor values versus OEM.
I've spent a fair amount of time running these through online calculators, etc and have only managed to confuse and concern myself further, so that's why I'm asking for someone with more experience to decipher the OEM design and let me know if it is correct as-is, and explain why, or if it doesn't seem right how best to correct it.
What's interesting is when I run this through a 2 way crossover design calculator, simply modeling the tweeter and mid-range at the 2500Hz point without considering the existing series resistors, the capacitor value for the tweeter and coil/inductor value for the mid-range comes out close (capacitor the same, .25 versus .5 mH coil/inductor) to what is currently in the circuit OEM. But not exact. See attached.
However a similar experiment choosing 300, 400, 500, or 600Hz as the crossover point for the mid-range and woofer do not generate the same capacitor and coil/inductor values that we see in the OEM circuit.
Of course modeling two 2-way scenarios isn't the same as the 3-way model, but it is interesting to note.
So ignoring the unknown sensitivity of the OEM drivers for a moment, in theory I should be able to simply change the 4 uF tweeter capacitor for a 8 uf tweeter capacitor to support the new 8 Ohm dome tweeter.
But again, that possible frequency "gap" between the tweeter and mid-range speakers that seems apparent in the crossover calculators has me asking for second set of experienced eyes to make sure that is all that is needed or recommended.
Thanks
Last edited:
Without actual measurements, this is more or less cut and try.
The impedance and crossover part values are not accurate enough
for precise modeling. Certainly the series resistors must be included as
part of the impedance.
The impedance and crossover part values are not accurate enough
for precise modeling. Certainly the series resistors must be included as
part of the impedance.
Understood, so since resistance is not exactly the same as a speaker's impedance, any thought on a typical ratio of the resistor's Ohm value to apply in estimating the impact to speaker impedance?
Much like when estimating a speaker's impedance with a multi-meter, a reading of say 6.5 Ohm would indicate an 8 Ohm speaker, would one apply perhaps 2/3 of a resistor's value to estimate it's impact to a speaker's impedance?
For example, the mid-range speakers have a 5.6 Ohm resistor, so what would be a safe value to use... 50%, 66%, of 5.6 Ohm.... then add that to the 4 Ohm speaker, or? I understand it's inexact, but what is a typical approach?