I recently had my crossovers rebuilt. We changed the tweeter resistor from 2.5 ohm to 3.9 ohm to lower the output. It dropped it about 2 dB per REW test. I would like to hear it again at 2.5, but instead of removing the 3.9 (Path audio) resistor, I was going to add a 6.8 in parallel, which would drop the resistance to 2.5.
But that value is on back order from Parts Conn, could I use a different brand, say Mills MRA 12 6.8?
2 questions, is mixing brands a bad idea, and mixing values?
Thanks
But that value is on back order from Parts Conn, could I use a different brand, say Mills MRA 12 6.8?
2 questions, is mixing brands a bad idea, and mixing values?
Thanks
Mixing brands or mixing values is not a problem. Mixing types also doesn't need to be a problem. What you should be aware of is the ratings and properties of the resistors.
Casually speaking, if they're the same type and if this is for crossover use then there is no big issue, simply test them in use to see they aren't getting too hot... For critical design you would at least consider (calculate/measure) each resistor separately in its application.
Casually speaking, if they're the same type and if this is for crossover use then there is no big issue, simply test them in use to see they aren't getting too hot... For critical design you would at least consider (calculate/measure) each resistor separately in its application.
Thanks. The Mills are nice and inexpensive, and it is likely a short term test. Also, I listen around the 70 dB range and my spks are 95 dB efficient, so there shouldn't been any heat issues.
I am a bit of a noob in this area, by type do you mean wire wound or not? They are both non inductive and 1% tolerance.
Ok, then I think we can all relate to the idea of short term testing.
Resistors have all kinds of properties.. power, Voltage, temperature coefficient, non-linearity, ability to withstand being pushed.. to name a few. Most of it doesn't matter for a crossover. Inductance is not a very important property for a crossover resistor. Tolerance can be compensated for.
Resistors have all kinds of properties.. power, Voltage, temperature coefficient, non-linearity, ability to withstand being pushed.. to name a few. Most of it doesn't matter for a crossover. Inductance is not a very important property for a crossover resistor. Tolerance can be compensated for.
99% of resistors contain iron, at least legs made out of it, also endcaps in film resistors.
Finding ones which do not, will turn a normal parts order from a known supplier int a very hard quest.
Just sayin´
OK, thank you for all your input.
Adding the 2nd resistor in parallel, which brought the (overall) resistor value back to spec, sounds better to me. Not the tweeter level, but it restored the crossover point to spec. I trust the person who rebuilt my crossover (he is currently building me a 2a3 SET amp), he felt the difference in resistor value (original change 2.5 -> 3.9 ohm) made about a 100 HZ shift in crossover point, but I think it might have been greater than that.
So I have a 3.9 ohm and 6.8 ohm resistor in parallel, which equals very close to 2.5 ohm (spec), in series with the 4 ohm tweeter. I was thinking of ways to drop the tweeter level, w/o changing the crossover points. I then realized the 3.9 ohm resistor and 4 ohm tweeter were more or less the same resistance, could I just move the 6.8 ohm resistor from parallel to the other resistor, to parallel with the tweeter? That should have the same overall resistance, and not affect the crossover point, while dropping the tweeter output level.
But, before trying that, I see 2 potential issues. 1st it will probably drop the output more than I want, I will take REW readings before and after, it will be interesting to see. But 2nd and main concern, it is a 2nd order crossover, so there is an inductor in parallel with the tweeter already. Moving the resistor will change the effective ohms of the tweeter from 4 to ~2.5. Will that affect how the inductor interacts, and mess up the crossover in a difference way?
I am still real new to this, so sorry if what I propose is totally wrong.
Thanks again
Adding the 2nd resistor in parallel, which brought the (overall) resistor value back to spec, sounds better to me. Not the tweeter level, but it restored the crossover point to spec. I trust the person who rebuilt my crossover (he is currently building me a 2a3 SET amp), he felt the difference in resistor value (original change 2.5 -> 3.9 ohm) made about a 100 HZ shift in crossover point, but I think it might have been greater than that.
So I have a 3.9 ohm and 6.8 ohm resistor in parallel, which equals very close to 2.5 ohm (spec), in series with the 4 ohm tweeter. I was thinking of ways to drop the tweeter level, w/o changing the crossover points. I then realized the 3.9 ohm resistor and 4 ohm tweeter were more or less the same resistance, could I just move the 6.8 ohm resistor from parallel to the other resistor, to parallel with the tweeter? That should have the same overall resistance, and not affect the crossover point, while dropping the tweeter output level.
But, before trying that, I see 2 potential issues. 1st it will probably drop the output more than I want, I will take REW readings before and after, it will be interesting to see. But 2nd and main concern, it is a 2nd order crossover, so there is an inductor in parallel with the tweeter already. Moving the resistor will change the effective ohms of the tweeter from 4 to ~2.5. Will that affect how the inductor interacts, and mess up the crossover in a difference way?
I am still real new to this, so sorry if what I propose is totally wrong.
Thanks again
You can't. If the woofer filter stays the same, every change in tweeter level means also a shift in the crossover point. However if the change in level is minimal, then the shift in crossover point should be negligible. Note also that a resistor in series with the tweeter won't drop the level equally on all frequencies, this is because the tweeter has an impedance that varies with frequency.
Please draw your current crossover, and the proposed mod. There isn't a single way to do a filter.
Ralf
It is easier for me to just show a picture. This is the old crossover, the new one with foil inductors is harder to see/understand.
"A" is the resistor in series with the tweeter, it is 2.5 ohm.
"B" is inductor parallel with the tweeter
"C" is tweeter connection, black because it is wired out of phase
"D" is the ground, connected to main negative speaker term and all speakers.
I know that is likely obvious, especially C & D.
The circles are not relevant here.
Darn in reply to your 1st comment.
Looking at this further, the inductor is in parallel with the tweeter and the resistor, so moving the 6.8 between the 2 shouldn't change anything. It may not change the tweeter level either though.
Not easy, but it seems a 3-way crossover, where the tweeter filter consist of a 2nd order electrical network followed by a resistor. In the attached image it is the upper filter. The other is what I think is your proposed mod.
Using the above components values and a Seas tweeter (or to be precise, its impedance), those are the predicted responses of the two filters:
Using the above components values and a Seas tweeter (or to be precise, its impedance), those are the predicted responses of the two filters:
Thank you. 🙂
The setup looks right to me, but R5 changes from 2.5 to 3.9 & 6.8 when it becomes R1/R2. R2 is correct at 6.8, R1 should be 3.9.
FWIW Tweeter is Satori TW29BN-B
I tried it, it dropped the tweeter about 4 dB.
But since all I did is move a 6.8 ohm resistor from parallel with a 3.9 (~4) ohm resistor to parallel with a ~4 ohm tweeter, all in the same path, the overall resistance should not have changed between points A and B. And if that value did not change, then the crossover points should not have changed.
Am I missing something?
I wouldn't expect much change, but then the tweeter impedance does vary with frequency.
You're not showing the tweeter response separately?
You're not showing the tweeter response separately?
Sorry, not quite sure what you mean.
That is a nearfield reading from about 1 meter. Left spk.
This driver is pretty flat in the relevant area. I have to make the change on the other speaker and see how it sounds.
This actually fits. The initial change from the default 2.5 ohm to 3.9 ohm resistor, dropped the tweeter about 2 dB. When I made the last change, that resistor (R1) went from 2.5 to 3.9 again, so I lost about 2 dB there and another 2 dB with the 6.8 added in parallel with the tweeter, for 4 dB loss in total.
I don't know how "tricky" this is, but I did manage to drop the tweeter without affecting the crossover frequencies. 🙂
I think with a 1st order crossover, this could be done without complication (assuming there was a resistor at all in series with the tweeter), and a 2nd order with this design, works also. The tweeter having a flat resistance curve is basically a necessity also. It occurs to me, other combinations other then 3.9 ohm and 6.8 ohm could be used, as long as the total resistance was 6.5 ohms in the end. If I wanted to drop the level less or more.