Improvements To Modified SDA SRS 1.2TL Crossover

ben62670

jcandy wrote: »

That's seems like a neat solution. Honestly that way you could more conveniently play with inductor (and series resistor) component values to optimize the woofer response.

I did mess with different resistors to see if the SDA was affected by the DCR of the inductor and resistors, but found the only thing that really changed was the bass slam. I was going to put a switch on it, but the series resistance had no beneficial properties to the music SQ.

DarqueKnight

OldmanSRS wrote: »

Is it accurate to say the SDA array is now being driven harder with stereo signal than is the stereo array below 150 - 50 Htz?

I think it is accurate to say that installing a lower DCR SDA inductor brings the speakers closer to their original design performance goals...hence Matthew Polk's statement:

"However, because of the DC resistance of the inductor, the system isn't perfect. I don't recall the DC resistance of these coils but it was high, at least several ohms depending on the model. This means that the SDA drivers will continue to produce some SDA signal even at very low frequencies. Decreasing the DC resistance will definitely improve the bass response of the system both qualitatively and quantitatively. "

I don't know if the SDA array is being driven "harder". It does not sound like it's being driven harder and I have not taken any measurements (yet). I did not hear any changes in image placement or sound stage dimensions.

OldmanSRS wrote: »

How does this effect the low/mid bass tonal balance and total bass output. It is a very positive change or is it unbalanced or excessive?

It was a very positive change for me, but it might not be to your liking depending on your room acoustics and personal preferences.

The result of this modification is more and better bass. It remedies a design compromise that was made to accommodate the amplifiers common at that time. If there were some high likelihood of unbalanced sound resulting, I think Matthew Polk would have specified other crossover adjustments to address that.

The following comments were made after the Northcreek inductor installation:

DarqueKnight wrote: »

Others have reported big improvements in the sound of the bass and I was expecting the same, but I only heard a modest improvement in bass detail and clarity. I felt a big change in the bass.

There was a big improvement in bass tactile sensation. When comparing the upgraded right speaker to the left speaker with the stock 16 mH coil, I could feel more vibration coming through the right armrest. I also heard more clean, defined rumble when I was in other rooms of the house.

The following comments were made after the Solen inductor installation:

DarqueKnight wrote: »

Right out of the box, the Solen inductors enhanced the improvements brought by the Northcreek inductors. There was significant improvement in center image weight, clarity and detail. Janne reported more driver excursion with the Northcreek 16 mH SDA inductors. I did not see this when comparing one speaker with a Northcreek inductor to the other speaker with the stock inductor. I saw (and felt ) more stereo driver excursion with the Solen inductor compared to the Northcreek inductor.

The only down side to this exercise is that I now hear more variation in recording quality, particularly in the bass region.

jcandy

Maybe this "cartoon" is useful for those that are confused by the SDA crossover. This captures the basic essence of the SDA network, while ignoring more subtle aspects.

Note that at high frequency, the 16mH inductors provide a nearly infinite impedance. Thus, they can be approximately removed from the circuit.

From this diagram, its easy to see why the left dimensional driver sees VL-VR and the right sees VR-VL at high frequency. Its also easy to see how at low frequency, the system reverts to a more conventional network. Note also that the actual dimensional drivers are closer to the centerline than the stereo drivers.

ben62670

With the IC disconnected the dim drivers act like subs. The lower picture should still show the inductors between neg of the dim drivers and ground. The break in the circuit would be between the inductors with the IC removed.

jcandy

ben62670 wrote: »

With the IC disconnected the dim drivers act like subs. The lower picture should still show the inductors between neg of the dim drivers and ground. The break in the circuit would be between the inductors with the IC removed.

I can see that the picture may require clarification.

It is assumed that the IC is connected in both cases. The bottom is the low-frequency limit, for which the inductors have nearly zero impedance and are therefore just shorts.

You are right that the dimensional drivers will be passives (I assume you meant to say passive not subs) if the IC is not connected.

ben62670

Actually they act as subs because they still get the signal to the ground via the inductors. The inductor basically acts like a 6db lowpass.

DarqueKnight

Apparently, "confusion" about the SDA crossover is more widespread than initially thought.

jcandy wrote: »

Maybe this "cartoon" is useful for those that are confused by the SDA crossover.

ben62670 wrote: »

With the IC disconnected the dim drivers act like subs. The lower picture should still show the inductors between neg of the dim drivers and ground. The break in the circuit would be between the inductors with the IC removed.

jcandy wrote: »

I can see that the picture may require clarification.

You are right that the dimensional drivers will be passives (I assume you meant to say passive not subs) if the IC is not connected.

ben62670 wrote: »

Actually they act as subs because they still get the signal to the ground via the inductors. The inductor basically acts like a 6db lowpass.

jcandy wrote: »

These are massive air-core inductors (used to remove the SDA effect at low frequency), and with large air-core units there is a very large DCR.

DarqueKnight wrote: »

As I mentioned in post #57 of this thread, I actually measured lower DCR with larger air-core inductors:

Stock 16 mH, 18 AWG, SDA inductor DCR: 2.8 Ohms.
Northcreek 16 mH, 14 AWG, SDA inductor DCR: 1.3 Ohms.
Solen 16 mH, 10 AWG, SDA inductor DCR: 0.8 Ohm, measured with VOM, 0.62 Ohm measured with woofer tester software (spec is 0.56 Ohm).

Such good circuit theory.

jcandy

ben62670 wrote: »

Actually they act as subs because they still get the signal to the ground via the inductors. The inductor basically acts like a 6db lowpass.

Ah yes, subs, obviously. Anyhow, at the risk of misleading people I think the diagrams were really supposed to indicate what's happening at low and high frequency respectively. Its also shows clearly why the inductors are parallel, which is what I was looking for when I made the sketch this afternoon. I could make three pictures, (1) HF, (2) LF, (3) no IC, which add more "realism" if anyone is interested.

jcandy

DarqueKnight wrote: »

Apparently, "confusion" about the SDA crossover is more widespread than initially thought.

Holy cow. You will note the diagrams are correct.

DarqueKnight

jcandy wrote: »

Holy cow. You will note the diagrams are correct.

Oh, OK, in that case, "my bad". Carry on. This circuit theory stuff makes my head spin.:frown:

jcandy

DarqueKnight

As I mentioned in post #57 of this thread, I actually measured lower DCR with larger air-core inductors:

Stock 16 mH, 18 AWG, SDA inductor DCR: 2.8 Ohms.
Northcreek 16 mH, 14 AWG, SDA inductor DCR: 1.3 Ohms.
Solen 16 mH, 10 AWG, SDA inductor DCR: 0.8 Ohm, measured with VOM, 0.62 Ohm measured with woofer tester software (spec is 0.56 Ohm).

I don't know why you reposted this. The point is that air core inductors at the same cost or gauge have much larger DCR than steel-laminate/ferrite core. It is for this reason that one generally avoids air core for large inductances.

jcandy

DarqueKnight wrote: »

Oh, OK, in that case, "my bad". Carry on. This circuit theory stuff makes my head spin.:frown:

I'm not trying to confuse anyone. I wanted to come up with something really simple (even bordering on too simple) to illustrate how the SDA crossover works, removing everything that is superfluous. I honestly hope it helps people.

ben62670

These are the coils that I have used with great results.
http://www.parts-express.com/erse-16-gauge-inductors.cfm

DarqueKnight

jcandy wrote: »

I don't know why you reposted this.

Just funnin' with ye. Relax. This is all very interesting stuff.

jcandy wrote: »

I'm not trying to confuse anyone. I honestly hope it helps people.

I know you mean well. Carry on.

ben62670

jcandy wrote: »

I'm not trying to confuse anyone. I wanted to come up with something really simple (even bordering on too simple) to illustrate how the SDA crossover works, removing everything that is superfluous. I honestly hope it helps people.

It took me a while to understand how simple it was because it looked too simple, and people were saying how complicated SDA was. The stereo and dimensional XO's are both simple 12db xo's that cross at the same frequency. The value difference is because one array(stereo) is 8ohms, and the other is 4ohms. After the positive passes through the dimensional driver it is sent to the other channels dimensional's negative side. When the signal passes through the dims ground it is out of phase. This is where you get the crosstalk cancellation. The inductors just prevent the low frequency signals from canceling each other out.

jcandy

ben62670 wrote: »

It took me a while to understand how simple it was because it looked too simple, and people were saying how complicated SDA was. The stereo and dimensional XO's are both simple 12db xo's that cross at the same frequency. The value difference is because one array(stereo) is 8ohms, and the other is 4ohms. After the positive passes through the dimensional driver it is sent to the other channels dimensional's negative side. When the signal passes through the dims ground it is out of phase. This is where you get the crosstalk cancellation. The inductors just prevent the low frequency signals from canceling each other out.

Agreed. I have this kind of saying that everything is simple ... once you understand it (with the exception of women and taxes). You echo that sentiment when you say "It took me a while to understand how simple it was".

ben62670

When I lived in Ft Lauderdale I was walking down the beach and I stumbled upon a lamp. I picked it up and rubbed it. Sure enough a genie came out and said he could grant me a wish(Genies only grant one wish not three). So I just had my boat stolen and I was going to go to the Bahama's in it. Instead of another boat I asked the Genie to build a bridge across the 60 mile span. He said it was too far and the seas were to deep. He then asked me if I wanted anything else. I said I would like to be able to understand women. He then replied "how many lanes do you want on that bridge".

OldmanSRS

ben62670 wrote: »

These are the coils that I have used with great results.
http://www.parts-express.com/erse-16-gauge-inductors.cfm

For $28 each, I will give these a test drive. Using the 17 mH coil because the 1.2's use a 16.8 mH inductor.

ben62670

If you are not satisfied you can return them. I am sure that you will like them. Also solder on the very ends so you can snip just a little just in case you do want to return them.

thejck

Forgive me for beating a dead horse if that's the case.
I know that DK tried replacing and sent back the 2.0mH and 1.0mH inductor on the midbass section of the 1.2tl circuit to NorthCreek.
Is that still the case or have other inductors been tried in their place. Is there a recommendation or are the stock inductors still the ones to stick with.

OldmanSRS

Please don't take offense to this question, but what exactly is the point of changing the inductor in the DA XO to a lower DCR? Better sound by fixing a Polk design compromise mandated by the low impedance drive shortcomings of vintage amps? Better sound through improved damping factor?

ben62670 wrote: »

The value difference is because one array(stereo) is 8ohms, and the other is 4ohms.

This supports the statement I made earlier how changing the DA inductor DCR will drive the DA harder than the SA at low frequencies.

As designed the series/parallel MW DA is 4 Ohms and the drive current passes through a 2.8 DCR inductor and a .8 DCR inductor for a total load of 4+2.8+0.8 = 7.6 Ohm (Z+DCR)

The MW SA series/parallel network is 8 Ohm and the drive current passes though a 1.2 DCR for a total load of 8+1.2 = 9.2 Ohm (Z+DCR)

As designed the DA is already getting slightly more drive at low frequencies than the SA based on 7.6 Ohm (DA) vs 9.2 Ohm (SA). You can visualize this by driving the speaker at 25 Hz and compare the cone excursion between SA and DA.

Changing the the DA inductor from 2.8 DCR to 0.5 DCR (Ertse) yields 4+0.5+0.8 = 5.3 (Z+DCR), about a 42% less resistive path or 42% more drive current. This should indeed yield more bass and improved damping but is this keeping with the true design and performance of the speaker?

Referencing my original question...

An approach that would keep the design true to the original and the way it probably should have been done by Polk would be to change the inductor AND change the DA MW's to 6503's. This would increase the DA's MW Z to 8 Ohms yielding a total of 8+0.5+0.8 = 9.6 (Z+DCR) which is a perfect match to the SA's MW 9.2 Ohm (Z+DCR). The damping factor would improve, the total load to the amp would still be around 4 Ohms and within the comfort zone of most amps, even the vintage ones.

My .02
Greg

ben62670

The low DCR of the 16mh inductor just improves the bass. It doesn't effect anything else IMO. Keep in mind that the signal the amp receives isn't that much lower in impedance because the signal still has to pass through the dimensional speakers before it gets to the inductors. Polk used aircore inductors back then and didn't use any iron cores. The amps are only taxed the resistance drop on low bass passages, and are not affected at frequecies above the inductors cut off point.

OldmanSRS

ben62670 wrote: »

The low DCR of the 16mh inductor just improves the bass.

Yes, by driving the DA 42% harder than it is currently driven and harder than the SA is currently driven at bass frequencies. That's the essence of the point I was making. I like to do the math....

I'm not suggesting it won't work and increase the bass and damping factor and has no effect of dimension or performance above about 150 Hz. I want to try it and hear the results for myself. It might just be one those must have mods.

ben62670

I know it made a huge improvement in mine. The overall impedance the amp sees will still be relatively close to stock. If your amp is rated at 4ohms then you should have no problems. What amp are you using BTW?

OldmanSRS

Carver TFM-35 SPECIFICATIONS
Frequency Repsonse: 20-20kHz +0,-0.2dB
Power Output: FTC 20hz-20khz
8ohm 250 watts per channel
4ohm 400 watts per channel
2ohm 750 watts per channel
Bridged mono power output: 800 watts into 8 ohms
THD: less than 0.08%
Dynamic Headroom: >1.0dB @ 8ohms both channels driven
Gain: 29.0 dB (+/- 0.5dB)
Input Impedance 47 kohms
Damping Factor: >150 typical (20hz-20 khz)

ben62670

No Problem You are going to love it.

jcandy

OldmanSRS wrote: »

Yes, by driving the DA 42% harder than it is currently driven and harder than the SA is currently driven at bass frequencies. That's the essence of the point I was making. I like to do the math...

You are spot-on with your observations. As I have said repeatedly, one will want to be very careful with this modification. Personally, I would not do it without some attempt at measurement of the response, and ben62670's approach of combining and separating the large inductors seems like a good idea so that you could add a series resistance "to taste", as I see no reason why minimizing the DCR must necessarily be better. If this very significant modification actually improves the SQ, it implies that the Polk engineers did a rather poor job at balancing and tuning the original design (which seems unlikely).

ben62670

It's not that the Polk engineers did a poor job of design. The speakers dip fairly low so I assume, and have had confirmed by Matt Polk that the extra resistance of the coil is to make the speaker is easier to drive. Amps that can handle dips below 4ohms can easily take advantage of this mod.

DarqueKnight

ben62670 wrote: »

You are going to love it.

I do love it!

One of the surprising things was the increase in low level listening pleasure. I can now listen at lower volume levels without giving up a large amount of musical detail. This is important for background music at home and even more important for background music while at work.

I was hesitant to do the inductor mod on the pair of CRS+'s I use at work because I have to listen at such low levels and I didn't think I would hear the benefits of the modification. I took one of my previouly modded CRS+ pairs to work and found out I was wrong. The increase in sonic weight, overall clarity and particularly bass weight and bass articulation was immediately apparent, even at the low volume levels I listen to at work and even with the modest electronics in my work audio system:

Nakamichi CA-5AII Preamplifier
Yamaha TX-1000U Digital Tuner
Yamaha CDX-1110U CD Player
Adcom GFA-555 Mk II Power Amplifier
Signal Cable Analog 2 Interconnects
Monster Cable Z2 Reference Speaker Cable
Signal Cable Power Cords (Pre and Power Amps)

Reference: SDA CRS+ SDA Inductor and PCB Mod

OldmanSRS

I can understand that. Driving the DA 42% more is adding more bass, like a loudness control. My main concern while I decide if I want to do this mod is what will the bass be like at moderate to loud levels. I can't get my mind around if adding more output in the 15 Hz to about 2 Khz range will make the speaker better. It will certainly change the bass/midrange/treble balance somewhat from the original sound.

DK, you think the positive benefit is still there when you drive them hard?