Autotransformer question.

[longdrive03]

I've got some mid compression drivers that are 114db and the 12db attenuation of the T2A autoformer isn't enough to pad down to 96-97db. Can you connect two autoformers to get additional attenuation of 18db? Can you use a padding resistor with the autoformer and if so should it be placed after the autoformer? Thanks.

[Guest " "]

I've got some mid compression drivers that are 114db and the 12db attenuation of the T2A autoformer isn't enough to pad down to 96-97db.  Can you connect two autoformers to get additional attenuation of 18db?  Can you use a padding resistor with the autoformer and if so should it be placed after the autoformer?  Thanks.

you need a 28db reduction and running 2 in series will only provide a 24 db reduction.

a lower cost alternative with better specs would be to use the below type of autoformers which are rated at 75 or 150 watts and have a frequency response of 20hz to 20khz +- .5db.  They have either a -35db or -43db  max reduction which means they can provide the approximate 256X multiplication factor you are looking for.

http://www.lightav.com/home/russound/volume/volume.html

russsound and a few other companies make these, so shop around.

[djk]

First off, what horn and driver is it?

Companies like BMS exaggerate their sensitivity specs by giving them on 40°x20° horns, not likely to be something you will use at home.

Horns you're likely to use at home will be more on the order of 110dB~112dB, and if they are 16 ohms they will be 3dB less than that with 2.83V/1M.

So if it is a 16 ohm driver and a very high gain horn at 112dB then a T2 will knock it down to 97dB/2.83V

[Erik Mandaville]

"Can you use a padding resistor with the autoformer and if so should it be placed after the autoformer?"

If you're not opposed to the technology, you can also use a two-resistor L-pad calculated for the required attenuation while maintaining the impedance of the driver.

Erik

[longdrive03]

The drivers are a TU 150 China made ("Design in USA) and are very large and heavy. They sound surprising good and clear when hooked up as the mid to my Heresy's I'm building for a friend. Definitely hotter than the Atlas K-55-G mid driver I'm using in the Heresy's which is 16 ohms but measures around 11 ohms. The TU 150 measures 15 ohms. I don't have test equipment but using the same autoformer taps on the Tu 150 it is clearly hotter. I may go with the 12db tap on the T2A and just add another woofer for the next set to bring up the bass. Looks like I would need 19-20 db padding if the 114db rating is valid. Here's a link to the driver off ebay.

I read the discussion about autoformer versus 2 resistor attenuation and it appears the autoformer gives much better response. Since I had a couple I used them. Works great on the K-55-G. Thanks for the response.

I

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=170136957204&ssPageName=STRK:MEWN:IT&ih=007

[DrWho]

I think Dennis is on the right track with this one...there is no information regarding how the SPL of the driver was determined. Without any means to measure I think you're shooting in the dark.

[LARRY]

TU-150

150W

16

180-6000Hz

110dB

43 OZ

145(D)*90(D)

3.60

This is the actual specs from the distributors web page. Response is 180-6000 hz (no +-) and 110 db. Hope this helps

Larry

[longdrive03]

Thanks Larry. Can you post the website link please? This thing weighs a ton! looks like I could go with the -12db tap on the T2A and be close enough for government work.

[LARRY]

http://www.dswbrand.com/pa_horn.php

Site for specs on Tu-150 Driver

[Erik Mandaville]

" I read the discussion about autoformer versus 2 resistor attenuation and it appears the autoformer gives much better response."

That wasn't my experience; rather the opposite, actually. I was comparing attenuation differences in a network I built using the resistors, and didn't expect them to sound necessarily 'better' or worse than an autoformer. Resistors work very well, and have been used by many companies for a long time. They are also used in the new Heritage edition rather than the autoformer. Paul Klipsch didn't care for them, but I suspect I have different preferences.

But absolutely! use what you think is best for you.

Erik

[longdrive03]

Thanks Erik. I just finished a set of H II's for my friends garage and will try the double resistor method on my next set. I've got two autoformers left and I'll compare them with the the padding resistors. Thanks.

[Erik Mandaville]

"I've got two autoformers left and I'll compare them with the the padding resistors."

I think it's always good to experiment, and it may be something you like. If not, you'll know yourself what works best for you. There is a little 'catch,' with this, though: The values of the other components in that section of the crossover were selected for use with the autoformer, not the L-pad. The autoformer alters the impedance of the driver whereas a series/parallel resistor L-pad is designed to maintain that impedance. I have done what you suggested in past, and it will *work*, however the performance I referred to above is based on a network specifically designed around the impedance of the driver.

In this case, since it had a determining factor in the original design of the Heresy II crossover, I think the end result will likely be in favor of the autoformer. I brought up the resistive L-pad device mainly in light of your original question.

Erik

[Erik Mandaville]

Also: Many here are so used to the concept of using a multi-tapped choke (autoformer) as an attenuation device that the single series or series/parallel resistor approach seems very foreign. Paul Klipsch was also one designer who preferred to use such chokes to balance driver outputs. However, it's a fact that autoformers in crossovers are much less common than resistors. I was building crossovers 12 years before I even came to this forum, and had not seen the autoformer approach until we bought our first pair of Klipsch Heritage speakers: La Scalas. I had known about their use as an impedance multiplier or impedance 'buffer' between an amp and speaker inputs, as well as input grid chokes in tube amps, but had not seen them used in crossovers. It was always variable L-pad tweeter and midrange controls, such as those used by companies like Altec Lansing, or fixed resistors.

So, there is a choice: One can design a network that transforms amplifier energy into heat while maintaining a more suitable load impedance, or one that goes about the business of output reduction in part by way of gross alterations of that impedance. 'Losses' are associated with both, which is why the end result is a reduction in driver output. My point is that resistor-based L-pads have been and continue to be, as can be seen with new Heritage line (not to mention the speakers designed and sold by dozens of other companies), an effective and appropriate way to balance drivers of differing efficiency. They are not inherently *bad* things, and in my experience do not compromise performance any more than autoformers -- this last point being subjective. Like ANY other aspect of this hobby, there is a choice, but being in the position to both make and implement that choice requires time and a willingness to learn, experiment, make mistakes, and hopefully learn from those mistakes.

Erik

[Erik Mandaville]

And Lastly:

If you know the impedance of the driver, you could also design a straightforward bandpass with a variable L-pad mounted on the back panel. You don't have 'hit or miss' or 'shoot in the dark' with any of this. If the driver is 16ohms, use a 16ohm variable L-pad and adjust the output until it blends the way you like it. That's one of the other problems with an autoformer like T2A: You are obliged to use the reductions imposed by the taps with nothing in between. The compromise doesn't have to do with performance, but with the fact that you would have to drill a hole in the back panel. Alternatively, you can install the L-pad on the crossover board itself.

If you were interested in trying this, I have two essentially new L-pads, for 16 ohms, that you can have. They are the ones used in these networks I built for a K-horns a few years ago. This wasn't just an autoformer/L-pad swap; the network was designed from the ground up. Anyway, I'm not using them and you're welcome to them if you want to experiment. The L-pad shown below was taken apart so I could drill a small hole in the middle of the housing so it could be secured to the board with a small screw.

[longdrive03]

Gotcha. I think! I'll try the padding resistors and calculate the xover values at the proper frequencies based on the driver impedance at those frequencies and give it a go. They are a lot cheaper than the autoformer. I live just down the road from Bob Crites so I can get some quickly if need be. Thanks again, I appreciate all your input.

[longdrive03]

I did a sweep on my woofer tester II and this TU 150 driver has high impedance. At 742 Hz it is 17.5 ohms and at 6057Hz it is 36.52 ohm. at 1255 Hz it is 78.6 Hz???. I don't have the equipment to do a frequency response graph (or the know how) but this sure looks different than the K-67 driver impedance values I measured. I'm concerned about the response of the TU150.

[Erik Mandaville]

Sure, you can give the L-pad a shot, and if you're not happy with it, the autoformer will always be there.

With the climbing impedance, it might be helpful to install an RC network across the driver. You already know the DC resistance, but do you know the value of inductance of the voice coil?

Erik

[longdrive03]

Yes, the sweep shows .14mh at 742Hz and it rises to .64mh at 6057Hz with a high of 6.38mn at 1255hz which corresponds with the 78 ohm impedance.