Deang Posted December 7, 2004 Author Share Posted December 7, 2004 That's pretty much it. I would have done a thread like this a long time ago, but there wasn't really any need since no one was asking. Now people are asking. I'm hoping the thread will cut down on my email, since I just don't have the time to answer it all. Chris' new boards will cut down on much of the work, and will make this cake work for anyone who wants to give it a shot. Of course, my position is that this mod offers more performance for the dollar than anything else out there. It's a complete no-brainer. I would have liked to show how to solder and use the braid, but I only have two hands. Maybe I can talk Debbie into taking some pictures for me when she has time. Bottom line: Work slow, and work clean. Quote Link to comment Share on other sites More sharing options...
Al Klappenberger Posted December 7, 2004 Share Posted December 7, 2004 You guys realize Dean did all this becasue I said he didn't know how to write assembly instructions! Now, let's see if I can think of something else I can con him into doing! Al K. Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 7, 2004 Share Posted December 7, 2004 ...and FWIW, I think he did a fine job capturing this effort on "film". Nice job, Dean. Can't wait to see what Al has in store for me when I release my stuff..... Popbumper Quote Link to comment Share on other sites More sharing options...
Deang Posted December 7, 2004 Author Share Posted December 7, 2004 Plenty of typos -- had to go back and fix 'em. I probably avoid the writing because it's a big part of my day job now. I have to write a System Requirements Document for each network I design, and usually come home with my eyes glued half shut and a bit of headache. I took off work today -- and decided to trade one kind of headache for another. Chris, I'm sure you've already figured out that if you make the board much bigger, it won't be able to attach to the back of the terminal cup and pass through the opening. It's easy enough to remove the woofer and use that opening -- and mount somewhere in the cabinet. I was wondering if you've considered separating out the LF and HF circuits, and making two boards that can be stacked on top of each other like the RF-7? Mouser sells some nifty nylon spacers. You know what they say, it's cheaper to build up than to build out. Quote Link to comment Share on other sites More sharing options...
STL Posted December 7, 2004 Share Posted December 7, 2004 If we abandon mounting the PCB to the terminal cup (and I think we should) and go with two PCBs then why bother placing one of top of the other? If one is going to the trouble of doing the two boards (which I don't think we should do) then it would probably be best to mount the HF and LF sections away from each other to lessen coil/inductance interaction. BTW, nice job Dean! And the Radio Shack solder sucker does work fairly well -- you just have to clean it ever so often (and apply new lube to the moving seal). Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 7, 2004 Share Posted December 7, 2004 Funny part is, we are slow here at work right now and I have been doing sketches of the crossover mockup. DEAN: You must have been reading my mind (sure, it's like a "short story"), BUT, I am indeed finding with high quality components, the real estate gets eaten up quick. I HAVE pretty much abandoned the "make it fit like the original" concept. YES, I had considered the "dual board" solution, and it looks like this is going to be the best solution. Ideal? Well, problem is, if I DON'T, the board will end up being like 10" long and 4" wide. BAD MOVE. The idea here is to provide enough real estate so the end user can pretty much choose the largest components he/she wants, within reason. STL: Original drawings are pretty much aligning to a "split" concept. One board would have the Tweeter/Squaker section (already drawn), and the other board would have the Woofer section (and any Zobel network stuff, like on the Academy). Yes, I will need to work out connectivity between the two, OR, supply a cable that will split the input to both boards. Alright guys, anything else, LMK. It's in the works! Popbumper Quote Link to comment Share on other sites More sharing options...
STL Posted December 7, 2004 Share Posted December 7, 2004 popbumper, A few suggestions: 1) Keep in mind components can be mounted on both sides of a PCB (even if it's a single layer board). 2) Don't design the board for the extremes. By that I mean don't go too crazy trying to make room for extremely large caps that most people aren't likely to use. I suspect by scaling back so that it work with 80-90% of the quality caps on the market might drop the size consideribly. Those that want the crazy-big can make them work by using creative mounting techniques. 3) Note that some people might decide to use a cheaper poly cap (like a Dayton) in parallel with a fancy audiophile bypass cap (like a 0.1uF AudioCap) instead of one large audiophile cap -- so it might be nice if the board allowed for that. 4) I making the PCB universal I'm sure you've found cases where one design has a cap that others don't. I think it would be best to layout the board assuming there will be a cap and if there isn't the user can just solder some wire in place of the cap. 5) If you can get the PCB a little smaller where some might want to use it as a single board, then the layout could be done such that it's made as a single board and just has to be cut (with one straight rip) to make it dual PCBs. Quote Link to comment Share on other sites More sharing options...
STL Posted December 7, 2004 Share Posted December 7, 2004 Also, most all of what I said about caps can be applied to coils too. Also keep in minds that two air-core inductors should be mount differntly from one another to cut down on interaction. It's hard for me to describe with words, but one should be laid flat against the board (so looking through the hole you only see the board) and the other should be mounted on edge rotated 90 degrees so you can look through the hole. The coils should also be mounted as far away from each other as possible (and that probably goes for the autoformer too). Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 7, 2004 Share Posted December 7, 2004 Thanks for the inputs, my comments after "**": 1) Keep in mind components can be mounted on both sides of a PCB (even if it's a single layer board). **Absolutely, BUT, it makes for difficult mounting 2) Don't design the board for the extremes. By that I mean don't go too crazy trying to make room for extremely large caps that most people aren't likely to use. I suspect by scaling back so that it work with 80-90% of the quality caps on the market might drop the size consideribly. Those that want the crazy-big can make them work by using creative mounting techniques. **There will need to be SOME compromise here - I am trying to allow for larger components from the "get-go" rather than creative mount techniques. Dean does a lot of this because the real estate forces him to; I would think that folks would want "simple fit and form", which is part of the exercise. 3) Note that some people might decide to use a cheaper poly cap (like a Dayton) in parallel with a fancy audiophile bypass cap (like a 0.1uF AudioCap) instead of one large audiophile cap -- so it might be nice if the board allowed for that. **By using a large footprint on the board, then anything smaller is an easy fit - so this is "built in". 4) I making the PCB universal I'm sure you've found cases where one design has a cap that others don't. I think it would be best to layout the board assuming there will be a cap and if there isn't the user can just solder some wire in place of the cap. **Absolutely, already considered. The board has spaces for all possible combinations where one CAN put a component in if it is needed, and it be JUMPERED or OPEN if it is not. I will include instructions outlining this. 5) If you can get the PCB a little smaller where some might want to use it as a single board, then the layout could be done such that it's made as a single board and just has to be cut (with one straight rip) to make it dual PCBs. **Hmm, that would be a lot tougher to do. If I made two identical PCB's that you could configure to work in tandem, there would have to be a lot more options for jumpering and bypass, and I would HAVE to choose very specific components for fit. Popbumper Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 7, 2004 Share Posted December 7, 2004 STL: Also painfully aware of coil interaction, which works out REAL NICE if the boards are separated; HF and MF can easily fit one board, and LF can be seperate. Many, many possibilities; just trying to get it "right". popbumper Quote Link to comment Share on other sites More sharing options...
STL Posted December 7, 2004 Share Posted December 7, 2004 I think we're on the same page with most issues, but here are a couple follow-up comments: 3) Real-estate wise I already assumed it was covered, but I was talking about you allowing for them by having extra holes in the PCB for the bypass caps. 5) If the two boards are roughly the same width then it would just be a matter running two traces (from the input) to the "other" board. Those traces would be cut when the board are separated. I was also thinking manufacturing costs might be consideribly cheaper with the one board (that can be make into two) approach. Quote Link to comment Share on other sites More sharing options...
STL Posted December 7, 2004 Share Posted December 7, 2004 popbumper, I thought you might already know about coil interaction, but I didn't want to make that assumption since it could have lead to a potentially costly mistake. I am just trying to do my part to help you get it "right". Once you get a final design you could post it here for a "peer review". Quote Link to comment Share on other sites More sharing options...
Joe Shmoe Posted December 7, 2004 Share Posted December 7, 2004 AND... if we could get a review on the sound differences on the KLF-30 once it is up and running with new crossovers? Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 7, 2004 Share Posted December 7, 2004 STL: No, no, I WANT the feedback, so don't think I did not; I was just commenting that I was aware of the coil issues! Popbumper Quote Link to comment Share on other sites More sharing options...
Deang Posted December 7, 2004 Author Share Posted December 7, 2004 Look at the stock networks and the close proximity of the coils to one another. You might consider the strong and very likely possibilty that any mutual inductance resulting from the near placement of the coils to one another was accounted for, and factored in. Adjustments would have been made to the parts values to compensate for it. In other words -- it may have been modeled that way. Isn't that a pleasant thought? Add that one to the list of reasons why I don't like to mess with them. Quote Link to comment Share on other sites More sharing options...
mike stehr Posted December 7, 2004 Share Posted December 7, 2004 Use the solder sucker to get most the amount of the solder blob, then wick it clean. Otherwise, it takes six miles of solder wick to wick off a big clump of solder. Quote Link to comment Share on other sites More sharing options...
fini Posted December 7, 2004 Share Posted December 7, 2004 Mike, Didn't Tennessee Ernie Ford have a song about that? Nice work, Deano! Do you only answer emails if asked to in posts? Quote Link to comment Share on other sites More sharing options...
STL Posted December 8, 2004 Share Posted December 8, 2004 On 12/7/2004 6:44:06 PM DeanG wrote: Look at the stock networks and the close proximity of the coils to one another. You might consider the strong and very likely possibilty that any mutual inductance resulting from the near placement of the coils to one another was accounted for, and factored in. Adjustments would have been made to the parts values to compensate for it. In other words -- it may have been modeled that way. Isn't that a pleasant thought? Oh my -- you very well may be correct!! Quote Link to comment Share on other sites More sharing options...
popbumper Posted December 8, 2004 Share Posted December 8, 2004 Interestingly, I would not think that was the case - yes, I can see where they were "close" but I can't agree that there would be any substantial interaction and it was compensated for. Given that there is nobody to really "ask", I guess we'll have to stick with the original values and make best effort to keep things away from each other. Unless of course one wants to reconsider redesign of the whole thing - not me... Popbumper Quote Link to comment Share on other sites More sharing options...
Deang Posted December 8, 2004 Author Share Posted December 8, 2004 PCB: TYPE CEM-1 CU. WEIGHT: C: "AS INDICATED", TYPE: MYLAR L: "AS INDICATED" R: "AS INDICATED" MUTUAL INDUCTION: VALUES OF "L" ARE MEASURED ON CIRCUIT BOARD AFTER ASSEMBLY. OTHER: N/A COSMETIC LABEL: N/A CE: STICKER: N/A, PLUGS: N/A LABELING: PART NUMBER WITH REVISION MODEL NUMBER: VENDOR NUMBER: DATE CODE: SECONDARY COMPONENT ATTACHMENT: PARTS ARE SECURED TO PCB USING SILICONE AND WIRE TIES AS REQUIRED. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.