These just in from Southbound:

 

Thanks,Iggy! For your help in posting these pics.

I hope they help to validate the use of the terminator mod for those who have both installed it, and for those who still wonder about its purpose, or question its worth.

Southbound

 

Southbound thanks for all your info on mods and polks.

I just ordered a pair of DX-6's for the front of my new TL.

I plan to start there and thinking about the Kicker RMB8 after.

I have read a lot of the posts and I'm getting a little confused so if you could please straighten me out it would be appreciated.

To date is the RC Terminator mod what is best to add to the stock system with the polk's upfront?

Do you still use the 4.7ohm 5watt resistor or just disregard that and use the RC mod only?

Thanks a lot

 

WPF,
To help with your confusion...

The TERMINATOR MOD is something that you can and should add to ANY SPEAKER in ANY SYSTEM. I use it in my car(s) and in my home stereo. It helps clean up the sound of ANY system by terminating the speaker run in an impedance that the amplifier would rather see. Otherwise, there is minor degradation of the signal as it flows from the amp to the speaker as shown in these scope photos.

The 5watt resistor mod was STRICTLY FOR THE POLK DX6 SPEAKER IN THE STOCK TLS SYSTEM. It helps to knock down the tweeter level which is somewhat excessive in the stock system due to the EQ circuit used to boost the highs of the crappy Bose speakers. TO DATE... I have found 3.9ohm/5watt resistors to be the best choice for this use. A 4.7ohm might knock things down a bit too much. You can use any value between 2.7ohm/5watt and 4.7ohm/5watt OR no resistor at all, depending upon your particular tastes and your set of ears.

When you install the new Polks, add the terminator mod no matter what... unless you want to live with less than the optimum sound that the amp and speaker are capable of delivering. THEN if you find the high frequency response to be TOO BRIGHT, add the 3.9ohm/5watt resistors in series with the POLK tweeter to cut down its output. This will cut down the bite of the over emphasised highs due to the stock EQ unit.

Hope this helps.

GOOD LUCK! Southbound

 

Thanks Southbound for staightening me out.

I've got all my parts except I was only able to find 47000 instead of 50000 for the terminator cap. Will that be OK I should I locate the 50's?

 

wpf,
47,000pF is close enough. The circuit components are not that critical. The audio does not actually go through the components, rather the circuit indirectly helps out the audio by stabilizing the speaker line reflections and the amplifier load. SO TOLERANCE is not that critical. NTE electronics does sell the 50,000pF disc caps, though. You can find a local dealer through their website... but if you have .047uF caps in hand... I wouldn't bother.

It's a pain in the ass to perform a REAL "before and after mod" listening test. You have to install the mod and install the speaker and then take the speaker back out and detach the mod and reinstall the speaker several times for a listening test. But it's the only way to hear the improvements for yourself. OR YOU COULD JUST TRUST ME and the photos above, and add the mod, and just know that things sound better with it. That's what I would advise, unless you want to take the time and have the interest to really explore the A-B listening tests. It can be fun IF you have the time....

It's a lot easier to try the terminator mod out on your home stereo speakers. Just get some clip leads and clip the mod across the speaker inputs. Listen to a few minutes of a nice quality recording without the mod, then clip the leads across the RC terminator and across the speaker. Listen again to the SAME music selection. You can easily go back and forth as many times as you like. I'm betting that you'll be able to hear the difference blindfolded after doing this several times! Just don't expect to hear any difference by INSTANTLY clipping and unclipping the termination. It's not like switching between a different pair of speakers, where you hear an instant result. It's a subtle change that takes your ears and brain time to analyze and appreciate the cleaner more detailed sound.

Good luck!
Southbound

 

southbound,

I've been reading about using 'coils/inductors' lately as a passive low pass filter. My question is, when would u use something like a coil to cut high frequencies and when would u use a resistor to cut high frequencies in the tweeter like you are doing here for your polks?

Is it because you're merely 'reducing' the level that you're using the resistor here? If yes, can you do this at any frequency in theory? Say you wanted to cut down on bass, would you add a resistor of an appropriate value in series with the woofer?

thanks

 

miwa,
Speaker crossover networks can range from simple single order 6db/octave arrangements to complex third order 18db/octave setups.

What you don't want to confuse is the purpose of a crossover network vs. the purpose of an EQ circuit.

If it were not for the fact that the factory BOSE EQ circuit is virtually unaccesable, I wouldn't be doing any of the mods that I am doing to the Polk speaker. (except for the terminator mod)

Therefore, my mods are a simple compromise to cut down on the excessive high frequency output caused by the BOSE EQ circuit. The resistor in series with the tweeter cuts down the level being fed to it after the simple highpass capacitor that is mounted on the back of the speaker. This cuts ALL frequencies equally. The crossover cap is what determines the frequency of the point where signals begin to feed the tweeter. The tweeter couldn't handle the large low frequency signals, so the series cap is installed as a single order crossover from the factory. I am simply cutting the overall level down a bit more to the tweeter so that it doesn't get overdriven by the Bose EQ. If I had my way, I'd have a 1/3 octave graphic equalizer that gives you complete control of what frequency response is fed to the speaker, and none of these mods would be necessary! Since that's not easily accomplished here, I have just cut down the tweeter level. It's a compromise solution. However, once the attenuation resistor is installed, an additional rolloff cap can be added to begin rolling off additional high frequency signals from the tweeter. One wouldn't normally do EQ adjustments at the high power speaker level... but this stock amp is small enough in power to allow for it in this case. As I said, normally you would do all of your EQ adjustments back at the preamp line level stage of the amplifier.

Now to answer your exact question... inductors are typcially used in the crossover network in series with the "woofer". This is as you say a low pass component that will not pass the higher frequencies. So in simple single order crossovers, you put a choke in series with the woofer and a cap in series with the tweeter. This way, the woofer sees the low and midrange frequencies and the tweeter sees only the high frequency signals. The value of the cap and inductor would determine the "crossover" frequency points where they would become active. Typcial highend crossovers are usually third order 18db/octave networks. That means that they have three "reactive" components in their makeup. Both capacitors and inductors are reactive components that are frequency selective. The frequency with which they become active, depends upon the value of themselves as well as the value of the other reactive and resistive components involved as well as the reactive impedance of the speaker itself. Usually crossovers are made up of "PI" or "T" filters in front of the respective speaker. A resistor or variable potentiometer "Pot" is typically used in series with the tweeter for overall brightness control. This can either be switched in and out or turned up and down if its a Power Pot. This is what I have simply done with the POLK DX6 speaker operating from the Stock TLS BOSE system. And again, because the system is relatively low power, we can get away with a 5watt resistor and a cap to ground after that for addtional high frequency rolloff. Higher powered amplifier systems would require impractical values of the resistor and that is when you would need to design a more elaborate crossover. It would be made up of caps and coils that would not dissipate heat as the resistor does. Caps and coils have a "reactive impedance" instead of pure resistance. That means that they shift the phase of the voltage with respect to the current flowing through them. The current through a cap LEADS the voltage by 90 degrees, and the current through a coil LAGS the voltage by 90 degrees. Thus if you have a cap and a coil in series, there would be "resonant" frequency, which is a point where the reactance of both components are the same. At this point, since each shifts the phase 90 degrees in the opposite direction, there would be a 180 degree phase shift, resulting in a complete null. If you go through the AC power calculations, you will find that NO POWER is theoretically dissipated across caps and coils, due to their reactive phase shifting properties.

If you wanted to use a coil to rolloff the bass, it would be difficult to do. Crossovers are designed to feed the appropriate speaker with the proper frequency range with which it is best designed to reproduce. This crossover is in the midrange of the audio spectrum. That is its primary function. It separates the high frequencies from the low frequencies and feeds the higher ones to the tweeter, and the lower ones to the woofer. If you want to do addtional "filtering" of the highs or lows, that would typcially be done in the amplifier stages. The crossover is only good at separating the signals in the middle of the spectrum. They are not typically designed to also roll off bass or roll off extreme highs. This is something that the amplifer EQ circuit is designed to do. (although, due to the low powered amp, I created a way to roll off additional highs at the speaker itself in my DX6 mods) Putting a resistor in series with the midrange speaker would cut down ALL frequencies fed to that speaker... NOT just bass frequencies. And the resistor would have to be much more powerful to absorb the larger and longer duration low frequency signals. Again, this isn't the purpose of a crossover network. You want to avoid doing things at high power that you can take care of back at low voltage signal levels in the amp. You could, in theory, make up an arrangement of a resistor in series with the woofer and a coil shunting across the woofer that would accomplish rolling off some bass, but the values would be pretty prohibitive. Also, coils come with many different characteristics, depending upon whether they are wound on air cores or iron cores... depending upon how many turns and of what gauge wire, etc. This determines the "Q" of the coil, otherwise known as it's figure of merit" . This figure of merit determines how sharply the coil would come into play. The larger the Q, the sharper you can tune a circuit, but it also produces more ringing and undesirable artifacts. You are delving into a complex set of problems with your questions. You are better off to stick to the fundamentals of using a crossover for what it was intended, and an equalizer for what it is intended.

Hope some of this rambling has helped you and others to understand what and why I did what I did with the Polk DX6 mods. As well as some general information.

Later,
Southbound

 

miwa,
To further answer your questions...

You WOULDN'T want to use a coil to cut down high frequencies of the tweeter in a speaker system. You would only use it to separate the high and low frequencies near the midrange in order to split the frequencies between the woofer and the tweeter.

If you put a coil in series with the tweeter... Yes, you could choose a value that started to choke out the signal above a certain level (like say 10Khz). BUT ALL frequencies below that point would be passed freely to the tweeter INCLUDING EVERYTHING FROM DC TO 10Khz. Therefore, the bass frequencies would also be fed to the tweeter and that would be a bad thing. So how do you get rid of the bass signals from being passed to the tweeter? You put a CAP in series with it. Basically, if you used a coil, you would be working from the WRONG end of the response curve. Once you have a "HIGH PASS" cap in series with the tweeter, you might think that you could then add a "LOW PASS" coil in series with that to decrease the upper frequencies. While that might work at the UPPER tweeter frequencies, You would end up with a series LC "resonant" circuit as described above in my other post. This would result in a LARGE NOTCHOUT of the frequencies around the point where the cap and coil are equal in reactance. You would end up with a BIGH HOLE in your tweeter sound... how deep and how wide, would depend upon the Q of the coil. So again, that approach should be avoided.

If you wanted to roll off the bass with a series resistor like I'm doing with the tweeter, it could be done... But it would take a much larger wattage resistor, and you would also be cutting everything from DC to the crossover frequency (typically 3khz) down by the same amount. Therefore, you wouldn't really be lowering the bass... you'd be lowering the bass AND the midrange. The only way to accomplish just lowering the bass, would be to put a resistor in series with the woofer, and THEN ADD A COIL ACROSS THE WOOFER. Again, this would take a LARGE wattage resistor. It would be better if it could be done in the amplifier EQ stage. I don't really know of anyone complaining of too much bass in the TLS stock system, so the point is probably moot, and merely for educational purposes. The only reason I'm cutting down and/or rolling off the highs is because of the stock EQ pre-emphasis. I'd rather remove the pre-emphasis from the EQ. I tried bypassing the EQ as several members have done, but found that it also contained a gain stage that left the overall stock volume too low with it bypassed. If you are adding an aftermarket amp to make up for the loss of gain, I'D DEFINITELY ADVISE BYPASSING THE EQ and none of this messing with speaker response would be necessary! The stock EQ does have a low frequency rolloff built in it at around 30 Hz. This would protect the crappy bose speakers (or any other door speakers)from low frequency damage due to music content with HIGH subsonic content. This is probably a good thing...unless you're a heavy rap listener.

Hope this lesson has been helpful.

I didn't mean to write a book, but somehow I ALWAYS do

Southbound

 

hmm.. thanks southbound!! From your answer I gather:

1. you wanted an EQ effect to reduce 'all' frequencies and not just to cut off a piece of the highs.

2. its impractical to use a coil / cap combination because it will cause 'big hole' in sound.

3. you wouldn't be doing all of this if you had the choice of manipulating everything at an unamplified level.

This suddenly brings another question up. Southbound, I want to let you know that you do not 'owe' me answers so if you 'dont want to get into it', just say .. "don't do it, its not meant to be that way stop being cheap and just add another amp with crossover adjustments, miwa"

anyway, the second question is based on 2. I didn't have any problems with stock components since they're all gone from my system but I was plain curious of what you were doing. But what you said in 2., brought something to my attention about what I might be doing.

now, I have set of components that are being fed from my amp. I recently learned that you can set up coils/caps on this amp to do a tri-way setup. Hence all my research into coils/caps lately So I thought, oh , how convenient. Now I can add some bigger woofers to perform 'mid woofer' range instead of beating on my poor midrange in my components. Mind you, it does a pretty nice job already but its just lacking that low end oomph. I have a sub providing it, but its coming from the back. While I have solved most of the localization issues with phase shifting and frequency adjustment, I know in my mind that its coming from the back so it kinda messes me up. Plus, sometimes I like to turn the sub off but then I notice practically ALL the smooth lows are gone too. Amazing how much sound we hear at the low frequencies. Anyway, I figured I could add a set of small width 8" woofers sort of like the Kicker RMB8s that people have been putting in... but I'll be figuring out a way to put em up front. Then, I figured I would follow the amp's suggestion of hooking up a tri-way on the front channels with a coil for the woofers and a cap for the components.

Now, what worries me is this 'hole' you mentioned about. There is a passive crossover already on the component set. In theory, its made of these coils and caps that I am going to add too. They're just set to split at a higher frequency for their purposes right? What if I now add a cap for the components? Wouldn't that somehow trigger a hole too when it reaches the resonant frequency that matches the coils inside the component crossover? Would I be better off just adding the coil to the woofer and leaving the component side alone? The components are rated down to 60hz, but I really wished to only let them be responsible to say 100ish or 120ish and let the additional sub take over from there and below.

 

miwa,
I have a few errors to correct in my posts from last night. I was tired and was talking off of the top of my head and from typical designs that I had used in the past.

I mis-spoke about the effects of a LC (coil/cap) resonant circuit as used in the example in my above posts.

1. First, just to be clear, when you put a choke and a cap in SERIES there will be a frequency where the reactance of both are the same. At this frequency, the circuit will (in theory) appear as a COMPLETE SHORT. Again, how close to a dead short, and how steep and wide the slope to this apparent short circuit depends upon the "Q" of the components. Series resistance of the coil windings, for example, will introduce properties that make the ACTUAL resonance deviate from the theoretical perfect design. (albeit not by much)

2. When you put a choke and a cap in PARALLEL there will be a frequency where the reactance of both are the same. At this frequency, the circuit will (in theory) appear as an OPEN CIRCUIT with INFINTE IMPEDANCE. Again, how close to infinity, and how steep and wide the slope to this apparent open circuit depends upon the "Q" of the components. The same resistive factors would prohibit the ACTUAL circuit from becoming infinite in practice.

Now that you understand LC resonant circuits a bit more... TO CORRECT THE ERRORS FROM MY PREVIOUS POST... I am used to using a series LC resonant circuit as a "trap". If you start with a series resistor that the signal flows through, and then put a series LC ciruit to ground from the load side of the resistor, the resonant short circuit will shunt certain frequencies to ground and therefore "trap" them out. I was thinking this way when I said that the series LC circuit in front of a speaker would create a "notch" or "hole". JUST THE CONTRARY... Since we weren't talking about having the series LC circuit placed across the speaker to ground, instead we were talking about having it in complete series with the speaker. In this case, there would be a frequency at which the LC components would look like a short. Depending upon whether or not a resistor is also in series with the speaker, a "SPIKE" would be created at and around the resonant frequency. How large of a frequency spike would depend upon how large of a resistor is in series or if one is used at all, and of course the Q of the resonant circuit. You would have a spike regardless of a series resistor, because on one side of the frequency response, the coil is acting as a LOW PASS and on the other side, the cap is acting as a HIGH PASS, and when they meet, the circuit acts as an ALL PASS. A resistor in this configuration would just accentuate the effect. Also I previously mentioned that at the resonant frequency, the overall phase would be shifted 180 degrees... THAT WAS MY SECOND ERROR. (It was late at night) The phase across the coil and cap are each shifting the phase by 90 degrees in opposite directions. Therefore, with respect to each other, they are 180 degrees out of phase during most of their active frequency range... BUT, At the resonant frequency, they cancel and bring things back to center or 0 degrees. Resistors in the circuit also shift the phase back somewhere between 90 degrees and 0 degrees, towards center, at differenct points throughout the frequency band. This is why speakers and crossover networks present VERY complex impedance loads over the entire frequency band. AMPLIFIERS DO NOT LIKE THAT PHASE SWING! THAT IS ONE THING THAT MY TERMINATOR MOD HELPS TO COUNTERACT.

OK, now that we're back on the same page... I'll try to answer your questions again. I DON'T KNOW IT ALL, so I can't predict or calculate all of the variables. Sometimes trial and error is the best method. I just wanted to be sure to correct my own erroneous statements above, so that some other EE doesn't call me an idiot.

Your second post helped me to way better understand your thinking and your original question(s).

Yes, you were correct with your understanding in the above posts labled as summary 1, 2, 3. The only error was caused by my mistake in explaining configurations of resonant circuits. IN #2 of your response, it would still be improper to use a series cap and coil in front of a speaker because it would cause a "SPIKE" or "BOOST" at certain frequencies. (not a notch)

I was thinking that you were trying to find a better way to deal with the tweeter element of a speaker system... I now know that you are trying to deal with adding another actual woofer to better deal with the bass response. In that case you WOULD want to put a coil in series with the woofer. This is as I stated in some of my rants last night. FOR A SIMPLE SINGLE ORDER CROSSOVER, use a coil in series with the woofer to pass everything from DC to a certain pre-determined midrange frequency, and use a cap in series with the tweeter to pass only frequencies above this midrange point.

Now what you're trying to do is much more complicated than that. I must assume that the crossover network that you have installed is more complex than the simple example that I just described. IF it has coils involved, it's already at least a second order network. This provides some isolation between components. Without an exact diagram of the crossover, it's hard to imagine what's going on, and whether or not adding another choke in series with a speaker would cause a frequency SPIKE? In any case... I DON'T think you should mess with adding additional caps or coils to a sophisticated crossover network. This stock crossover is specifically designed to "cross" at the point where the speakers can best do their respective jobs. If you mess with that, in the wrong way, not only might you add a frequency boost somewhere, you might also introduce ringing and other artifacts. However, if you add a THIRD speaker, and feed it from a LARGE choke in series, that might work to give you more bass. While the stock components would still be getting signals down to 20hz (although they are only rated down to 60hz) the addition of a third woofer/speaker would have to overpower their low frequency response. Bypass the stock crossover and put a simple coil in series with this new woofer and see what happens. I dont' mean to bypass the crossover to the rest of your component system, just don't go through it for the new woofer.

To properly do what you want to do to raise the lower cutoff point for the component speakers, would probably require a redesign of the current crossover. However, as I scratch my head, adding a nonpolarized electrolytic cap in series from the amp output and placed before the crossover could raise the low frequency cutoff fed to the componenent speakers. BUT YOU MIGHT END UP WITH A RESONANT SPIKE AND ASSOCIATED RINGING, ETC. You wouldn't smoke anything doing this, but it may or may not work as expected due to the following crossover network characteristics.

BUT with the added woofer, if you drive it directly from the amplifier output and add a coil in series with it, you should have no ill effects. This would be the simple single order crossover that I have described above. Choose a coil that passes everything from DC to 400hz or whereever you think sounds best. This would be a trial and error selection as well. AS LONG AS THE AMP CAN HANDLE THE ADDED LOAD, it will buffer the existing component crossover network from this new speaker and you can properly choose a coil for low frequency use. You will just have to live with the component speakers being fed lower frequencies than you would like. OVERALL, I think the new woofer would overpower their response down at the low lows, and sound just fine... unless you crank things to the point that you can hear the midrange speakers break up from low low frequency distortion.....

LOTS TO THINK ABOUT, LOTS TO EXPERIMENT WITH!

Hope this helped a bit more... It's about all I can help you with.

GOOD LUCK!

 

heh .. heh .. heh .. yes.. what have I got myself into right? at any rate, thanks for the input and I think I'll try to keep things simple first and just add a coil on the woofer first. Whenever that will be ...

but yes, originally, I WAS curious only about your tweeter situation but then slowly it morphed to my situation. Maybe I should've started a new topic

thanks again, southbound!