ALKs, P-Traps, and Everything

[markus111]

Hello, this is my first post to this forum! I have a pair of mid 70's Khorns that I built ALK crossovers for. I was amazed at the detail that was revealed - like taking a blanket off of the speakers! Much improved bass as well. I would be very happy with the system except for one problem - a glare in the upper midrange. It is especially evident on female vocals and violin. My ALKs are made with ok components - alpha core inductors, original T2A, Solens bypassed with Dayton audiograde caps. I have seen some earlier posts regarding the use of p-traps to tame response in the 5-6khz region, but there were so many varying opinions that I came away with no clear solution. I'm kind of at a crossroads right now. Before I spend more money and add to my growing collection of unused crossover components, I'd like to ask the experts here about some recommendations. 1.) Do the new autoformers and Hovlands make a big enough difference over the mid-fi components I am currently using? 2.) Has anyone had success using the p-trap on khorns to tame that unruly upper midrange? Any help would be appreciated!

Mark

[Al Klappenberger]

Mark,

Don't bother with the P-trap. My network takes care of the 9 KHz glitch as part of it's design.

The midrange harshness is probably the K500 horn. There are two things you can do to improve it. First, cover the outside of the horn with damping material - Dynamat. Secondly - you also could lower the level to the squawker. If you had the 3619 transformer the x-4 setting is calmer than 2-5 where you probably have the t-2a set at now. The 0-4 setting used on the AA network is WAY too hot for my network. The best bet though is to replace the K500 horn with and Altec 811b or the one I hope to offer soon.

If you are not using Litz wire inductos for the .2 and .3 mHy coils, REPLACE them! Solid wire inductors are very lossy at 6 KHz.

Al K.

[markus111]

Hi Al,

Thanks for your comments. I damped the k400 horns with Dynamat, and this made a huge difference. I ended up with the midrange taps at 5--2. I was using a .33mH foil inductor from parts express to start, as I couldn't find the Litz inductors. I unwrapped several feet of the foil until I reached .30mH exactly. This seemed to make quite a difference in the sound. Should .03mh make such a difference? I'm just curious, as many caps are spec'd at +/- 10% If this is the case, I'll measure all my parts more carefully!

I also ended up putting an L-pad on the tweeter temporarily to see if this would make a difference. With the tweeter reduced about the 3db, I'm getting really close to the sound I want. These networks are really starting to shine!

Mark

[Al Klappenberger]

Mark,

the difference between .30 and .33 is major. .33 would have messed up the filter. I don't like the foil inductors. They are better than solid wire but not as good a Litz wire for this appliccation and they cost too much. The foil inductors have there lowest loss frequency up too high.

Some +-10% caps are actully closer to the right value than 10%. You just can't be sure. I think what really important is that they be matched left for right.

Al K.

[tigerwoodKhorns]

Fow what its worth,

I have not built my ALK's yet (parts are still in a box), but I have researched them heavily on this website. I have noticed that those who build a set not using Als specified parts either complain or end up tweaking and tweaking until the desired result is obtained. On the other hand, those that use the recommended components or buy from Al hook them up, adjust the squaker by trial and error, and then love them. I think penny wise / pound foolish comes into play here.

Chris

[markus111]

Chris - I think I'm beginning to agree with you on this. My main focus is on amp building - I've designed and built SET 45s, SET 2A3s, push-pull triodes, pentodes, you name it. While parts quality is definitely noticable in an amp, it's generally the circuit topology that gets you the most bang for the buck. I'm new to speaker tweaking, and I'm amazed at how much difference there is from one inductor type to another. I don't regret building the networks from scratch - I've learned a heck of a lot about my Khorns in the process. I may eventually end up buying the complete networks from Al. After all, he's done all the blood, sweat, and tears already, so I may as well lunch off of his pain It's like my mother use to say when she would bake a new recipe - bake it EXACTLY as written the first time, and then tweak it around to your liking after that.

Mark

[tigerwoodKhorns]

Mark,

I didn't mean anything as a cut down (not that you responded that way). If you check out this site, Al did tests on several inductors, including home made ones, before determining that the Solen Hepa Litz were the best. AL also went to great lengths on testing the circuit to determine that the Hovlands belong in the 2.2 uF position, that the inductors need to be in a certain orientation to prevent coupling, etc and designed the circuit to present a stable 8 ohm load to the amp.

From what I have read (I have not built mine yet) one of the reasons that these networks work so well is that they use all high quality components. The only thing that I would feel comfortable doing is maybe using higher quality caps like Auricapps instead of Solens (although I chose not to because they are expensive!! It adds up quick)and maybe using bannana connectors (for looks only).

Chris

[Erik Mandaville]

Hi, Mark

"It's like my mother use to say when she would bake a new recipe - bake it EXACTLY as written the first time, and then tweak it around to your liking after that."

Your mom was a genius, and I think this is very important advice -- especially in audio. As a DIYer building mostly SET amps, I'm curious about your designs.

I think there is an automatic response to tweak in the name of improvement; which although a respectable enough notion, does not establish a baseline for comparison. I have also built several Transcendent Sound products for other forum members here, and have been completely impressed with the level of performance achieved with inexpensive passive components. The primary concern is the circuit design and accurate values of capacitance, inductance, and resistance.

Great post!

Erik

[markus111]

Hi Erik,

Thanks for the compliment - my Mom's a darn good cook too. I can only hope to be as good a DIY'er some day. My most recent amp, and the one I'm enjoying the most is an SE 45. This is my second iteration of this amp. For the first one I built, I copied the Eric Barbour "single ended glory" design exactly. It was a very nice sounding amp, with monster soundstage and glorious midrange. However, it was seriously lacking in the bass department. In my second amp I switched to an SRPP front end, and put a lot of work into the power supply. I ended up doubling the power supply available current, and raising transformer voltage enough to use a choke loaded input with no peaking capacitor on power supply input. The effect this had on the bass response was amazing. My little 1 watt RMS (2 watt peak) amp can get my Khorns to rattle the rafters. I'm using PIO caps for coupling and power supply bypass. But again, I could use garden variety poly caps, and the sound would still be wonderful. I'm still amazed by how unforgiving these crossovers are on component types! But, that's why I'm here. I get to ask the resident experts about an area where I'm a complete novice, and I've been getting some great advice.

Mark