Jump to content


  • Posts

  • Joined

  • Last visited

  • Days Won


Posts posted by peniku8

  1. 47 minutes ago, maxmercy said:

    Interesting question, but I do not know what will happen sim v reality for this.  Will you have shaped and non-shaped cabs built to compare?


    No, too much work. The straight port version is not practical for me anyways. I have built an MDF prototype with a straight port, but that one was tuned to 30Hz instead. I figured that even the 12BG100 doesn't generate enough output to justify a tuning so low for my application. In addition to that, the port area on that one was a lot less and way too small.

    I will build two to be used for a live setup for weddings and smaller gigs. Either way, most 12" subs on the market are tuned to 50Hz or higher, which just doesn't cut it for me and most of the remaining subs which would fit my SPL and extension needs are bigger/heavier than just 40lbs.

    That said, the 12BG100 surprised me with its weight. 19lbs is quite a lot for a 12" pro woofer with a neo magnet imo.

    I've finished the CAD files and am currently working on the CNC code. I hope to have one measured in the next few weeks.

  2. Nobody knows? Alright, I'm just gonna build the cabs and see how it goes. It's just 70$ of wood for two cabs so w/e

    And even if the response is not ideal, these will be used with dsp amps exclusively so I can iron out some flaws at least.

    • Like 1
  3. 3 hours ago, Alexrmont said:

    I am planning on building six of these for a mobile stage I use. We currently have access to pk cx800 subs and stack them. By reversing polarity on the bottom one, we get more push forward. Is this something we can do with the skhorn, or does it not need that since these are horns as opposed to ported boxes. Thank you for all of the work you have put into this design. 

    I'm guessing that you're talking about a Cardioid setup here, but there is alot more to this than simply flipping the phase of a sub in a stack.

    There are many way to do this, but most common would be either stacking subs in two rows (needs more floor space and additional delay+measurement work) or physically turning every 3rd or 4th cab around so that its rear faces forward. You'd then flip the phase and delay the signal according to the depth of the cab. Delay can vary, so its best to just try different settings, measure and listen.

    Typically, the 2 row setup increases front output and rear rejection and the latter one I described just improves rear rejection. In short, yes, it can be done. It will require some work and knowledge thou and if you come here asking if this would work in the first place, it'll probably be your job of setting that stuff up and aquiring the necessary knowledge. But as Leonard Lauder once said:


    If I don't learn something every single day, it's a wasted day.

    Theres alot of crazy stuff you can do to array subs.

    • Thanks 1
  4. 8 hours ago, jay michael said:

    They are addicting! 2 is a nice start but 4 has a nice symmetry to it. 6 makes a good stacking height for your tops but is an odd number for amp channels.... better just make it 8 haha! 

    3 21DS115-8 in parallel per amp channel sounds like something I'd do. Their impedence minimum is pretty low and should be around 2 Ohm with 3 drivers in parallel.

    You could also get a point-symmetric setup that way. LCR. With a Skram beneath and on top of each main speaker if theres enough room. Put Skhorns in between for convenience.

  5. 5 minutes ago, dgage said:

    I doubt it would make any difference.  Bass waves are just too long and will go through that foam like it isn’t there.  For speakers, sure but for subs, don’t think you’d be able to measure a difference with the foam.

    Ricci did just that some time ago iirc. There was a meaningful difference even below 100Hz if I'm not mistaken. Remember that reflections will pass through the foam multiple times on all walls. I think it was even published as article here on data-bass.

  6. 1 hour ago, Boomer1950 said:

    Question:  Should I cover the internal walls with 1" acoustic foam?  (like the picture below) 

    acoustic foam.jpg

    Doing so will dampen resonances and make for a better frequency response at the cost of a fraction of a db of output. It's not always needed, but generally a good thing to do. Especially when maximum output is not a key factor. You can put the foam anywhere where it does not obstruct the airflow. With your port pretty much on all side walls.

    I used spray adhesive to glue the foam to the Skhorn and staple gunned the foam to my Z21. I personally hated working with spray adhesive.

    What are you doing with the PVC pipe? Will that be part of the port?



  7. 16 minutes ago, klipsch said:

    Ground and neutral will use the same bus bars back at the panel, unless it is not the main panel. Subpanels can be wired separately (which then converge back at the main). The main bus bars should be connected to the pole that's at least 8 feet in the ground. 

    To me it sounded like the adapter would be installed after the RCD. That would last as long as powering on the first electronic device...

  8. 2 hours ago, Tahoejmfc said:

    Currently I'm using 8 AWG class 2 for my short run of speaker cable from the amp. It is what i had in my shop for welding and I still need to do some cnc routing to mount the nuetrik connector flush to the back of the sub..

    Google Images wire gauge charts to get the proper wire size according to the amperage of the circuit required. Wire size gets smaller with a larger number. 

    Referring to the 120V 30 amp circuit feeding the amp... I used 10 AWG for a 7' run from my main panel 30 amp D-Square fuse to the L5-30 (120V 30A) Connection for the amp(Amazon has these connectors for cheap).

    It seems like here in the US, most amps run off of the 30 amp 120V L5-30 connection (correct me if I'm wrong please)

    A friend's theater has the same connection for when they host Tahoe Wormhole (Bass Heavy Whomp Whomp) with a PK system and it crushes using just that one connection for 6 double 18 subs. The remaining highs/mids of the system are powered by 4 different 120V 15 amp Edison plugs.

    I know you can get all sorts of adapters from a 240V 30 or 50amp welding connection on Amazon.. to convert down to a 30 am 120V by just grabbing one of the white, green and black leads. The white(neutral) and green(ground) are actually bonded together at the panel.. so you technically have 2 grounds and one hot (black) connection at the circuit breaker making 120V at the rated amperage of the circuit breaker that the black is connected to.  

    Keep in mind that music is not a continuous sine wave. If you're putting out 500W true long term average into a sub, that's already pushing it.

    When converting a circuit down in amperage, remember that an additional breaker is required.

    8 AWG is way over the top for any normal speker application (unless you have 100m+ of wire run maybe) and is only used in this case to keep the testing environment as linear and constant as possible to get accurate test results from the DUT. For usual PA applications I'm using either 11, 13 and 15 AWG, depending on what I'm powering.

    For when we set up the PA for not-so-small stuff, it is usual to have at least a 63A 3 phase 230V service.

    I have not come across Powerlock/Camlock yet.

    I don't know why you would link ground and neutral. That sounds like a horrible idea to me.

  9. 8 hours ago, Tahoejmfc said:

    I ended up pulling new circuits off of the main breaker panel for a L5-30 30amp 120V connection for the amp and another 15amp connection for the tops. Previously the Sub amp was connected to only a 15 amp circuit and I could tell it was somehow limiting the real output of the amp when turned up. ( I might be wrong here but I swear the subs drive better when the PL380 has a 30amp circuit dedicated to it.)

    If you didn't trip the breaker it was not the limiting factor. However, the wire gauges will be smaller on the 15A circuit (Or higher? I still can't figure out AWG. LESS COPPER!) and thus, voltage sag will act like a compressor on your signal in the earliest stage of your equipment chain possible. The smaller the wire, the more voltage sag you'll get. When I turn up my clone amp in my house in France it dims the lights. When I do so in my house in Germany, no voltage sag occurs. Maximum sustained power output of the amp is around 4KW for several seconds. 3KW long term. I saw the amp draw over 15A from the wall on the 230V circuit. Skhorn in a 3000cuft room. Loud!

    I'll be using 30m (100ft) of 8AWG speaker cable for heavy duty sub testing to mitigate cable losses (with the amp on a 230V 32A circuit).

    • Like 1
  10. 5 hours ago, SME said:

    Not exactly.  If you cascade two BW2 filters (if your DSP allows you to), you get an LR4 filter, but this is not the same thing as a BW4 filter.  The LR4 and BW4 both have the same 24 dB/octave slope below the transition regions, but the transition regions are shaped quite different and have different phase characteristics also.

    Sorry I think the post wasn‘t clear enough. I was trying to highlight the correlation in terminology of ‚order‘ and ‚db per octave‘, which is why I didn‘t mention filter type at all. I‘ve described cascading filters a few posts before so I didn‘t include that in this post again.

  11. 3 minutes ago, klipsch said:

    Those turned out great! 

    Thank you, I love every aspect of these. They sound incredibly good and since the driver is in a horizontal position, sitting on them is an experience lol!

    I can tune them 5Hz lower if I want to and that only takes about 30 seconds per cab to do.

    I revised the CAD files and overhauled some joints for the next batch. There are some visible seams which I didn't bother sanding and I will try to not use Kreg screws anymore.

    • Like 1
  12. 39 minutes ago, jay michael said:

    I do have something that I am not fully understanding.  How would you change a BW or LR filter between a 1st 2nd 3rd or 4th order filter?  When I look through my settings on my venu360 the options for crossovers just has the list of filters, BW LR etc with options for slopes, 12db 18db 24db etc but there is no settings for what order they are... am I missing something?


    It's just a different way of describing the same thing. Every 6db/octave is 1 order. 12db/oct is a 2nd order filter, 18db a 3rd order and so on.


    36 minutes ago, jay michael said:

    Could you tell us more about your bandpass horns? Having such a flat response from 30 all the way to 190 sounds really impressive!

    Well it's based on a bigger design I have not yet built, but I needed capable subs which could easily be handled by two people, so I went with a 320L cab which comes in at 63kg with casters. (The casters on the cab were a horrible idea, they resonate at 42Hz. I will build a caster board for future versions)

    The measurement was a quick 2m 2V groundplane (4 Ohm 21ds115) at 50°F/10°C outside in my backyard with no objects within 20m.
    The graphs shows the two required EQ points to EQ the response flat, but they should sound pretty neat inside a small to medium sized venue even without EQ. The two cascaded biquad HPQs form a BW4.

    When I've made up my mind whether to post the build thread on AVS or on here I'll do so. The thread will contain more in-depth views and info which I don't want to clutter the Skram thread with.
    I will however not release any plans for those. If you want some and live in the EU, shoot me a PM 😉



  13. I have liked a 120Hz crossover setting when recently testing out my own bandpass-horn designs. Took 2 PEQ points to get the response within ±1db from 30-190Hz and the EV tops I used to test them with didn't have such a good mid/upper bass (impulse) response. If you're running the system flat and use a low shelf to shape the response to your liking, this crossover point works really well with these types of subs. Now if you like using the subs' volume control to shape your signal, I'd suggest running a lower crossover or just not doing so at all. I don't like hot subs crossed over any higher than 70Hz.

    You could also try crossover shapes with an even smoother knee. Cascade two BW2 and you'll get a LR4. Cascade 4 BW1 and you'll get what my Chinese amp lists as "Bessel", but it is not. Bessel is in between BW and LR, thus forms a +1.5db point at the crossover frequency in theory. Cascading 4 BW1 filters will result in a 4th order filter with considerable overshoot, so you could overlap the responses of subs and mains more and run subs hotter while maintaining a natural slope. You can experiment a lot with it, but I'd recommend doing so in a controlled environment outside with measuring gear. Once you get a response you like, it's easier to integrate that into a room instead of fighting both the room and the speakers at the same time imo.

  14. On 2/8/2020 at 5:04 AM, Trdat said:

    Where I am going with this is lets put aside the transients for a bass cab or a subwoofer. If the above is true then by simply crossing over your PA cab with a sub woofer will improve transients to the PA cab right?



    So you're saying that the transient intermodulation distortion can be mitigated when feeding a speaker a smaller bandwidth signal? This should be true, but I can't attest that this effect is audible or if it's even a problem at all, as long as the speaker is being driven within its linear operation range. I don't know if this effect is even relevant for loudspeakers and I haven't managed to find much about this on the internet. I've only come across TIM measurements of amps. And slew rate of amps is typically not affecting audio quality, unless you're planning to do something wierd.

    I don't think it's possible to improve the other two aspects of the transient response (ringing and group delay) with a simple HPF.


    @SME I'd agree that a resonance of lower Q is more audible than a resonance of higher Q with the same gain, but I think that the resonance of higher Q will be more annoying for the listener. When I'm working on a mix I almost never use narrow additive EQ points, because those tend to sound unnatural (unless you're going for a specific stylistic effect).



  15. I'm in the process of designing a small vented PA cab for B&C's 12BG100, tuned to around 40Hz. The current design utilizes two straight rectangular ports with a combined port area of 120cm².

    Problem with the port is that the handle placement is so awkward, that I'd have to make the cab wider than I'd like to, which is why I'm planning to go with a shaped port. The fact that doing so would also lower port compression a little is a nice side-effect, but it's not my main concern here. I basically want the port to "bend around" the handle. Pics below.

    I'm simulating a rear loaded horn in hornresp to be able to be more flexible with the port input data, but I'm not sure if this is the correct way to approach this, or if there are better alternatives. I had to increase the chamber volume by 10% to match the simulated response of the vented cab.

    I sliced the port into 4 segments to model it as accurately as possible. The cab in the middle is far from done and needs a few modifications still, since I just quickly threw this idea together in Inventor to grab the measurements for hornresp. The sims are very similar between the straight port and the shaped one.

    Am I on the right track?


  16. 11 hours ago, klipsch said:

    Great details and response as usual. I appreciate the write-up.

    I need to work on integration more, but these skrams are awesome. Finally getting the chest pressure with kick drums I was lacking. I need to try things with 3 blocked ports. If that works well, I am wondering if I just sell the sealed boxes and go all skrams... 


    I'm using the SKHorn in 1 port mode in my HT. I get no usable output below 18Hz (chuffing related), which is fine because I'm sitting on a tiled basement floor. Only TR I can get is from my mini riser.

    9 hours ago, Tahoejmfc said:

    What kind of gasket tape?

    Yah, Gorilla glue, Im not a wood worker, but more of a mechanical engineer so I don't really know what is the correct method for assembling wood, it was a learning experience....

    - sent from my vibrating keyboard with occasional shit falling off the walls of my garage

    Use regular oval-head screws. I use those black ones commonly used for sheet metal, which work great on plywood too. There are gaskets for the handles which are less than a dollar (penn elcom) but you can also use 20x2mm gasket tape like I did. I like the screws, they make for a clean look.


    • Like 1
  17. And btw, last time I manually routed an MDF sheet, I completed a full depth cut (1/2") for a 12" driver in probably less than 4 seconds at 20k rpm with a quarter inch bit. I've seen feed rates in MDF being up to 4x as high as in plywood, it cuts like butter. 

    I have a 4" (two blade) tool for the Moulder, which the manufacturer recommends running at 9k RPM. When running at 2k, you'll get severe backlash and I literally had smaller workpieces fly through the shop because of low RPM. 

    So in essence: start with high RPM and see what feed rate works best for you. Then lower the RPM until dust turns into small pellets and the wood doesn't burn anymore. 

    Did you make plans on how to do the assembly already? Glue and clamp? Kreg screws (those will not work well in 1/2" MDF, I didn't catch what thickness you got)?

  18. While you might burn the wood and dull the tool with too much generated heat (when moving too slowly for the set RPM), it's more likely that you get backlash when feeding too quickly with low RPM, which is more dangerous in my eyes. A firm grip on the router is important and imo the tooling should allow for almost any feed rate. 

    RPM and feed rate  and their relation differ with material, tooling (diameter and shape, amount of blades) and eventually the desired surface finish. There are great feed rate/rpm calculators, which should get you a rough idea on your settings.

    The chips should not be dust, it should be small grains.

  19. 16 hours ago, SME said:

    Thanks for helping to clarify things.  Though thinking about it now, when cutting circles I think you still want to cutter to enter "from behind" on the outside edge where there's slightly more material to be removed.  Another important detail when cutting circles is that the inside piece that's to be removed must be separately anchored, and directly clamping it is not usually possible.  Instead before making the cut, it should be anchored to a larger underlying sacrificial piece using a couple finishing nails, and the  sacrificial piece should be well clamped.

    As an advanced technique, there are very exceptional cases to the rule of avoiding climb routing: specifically when exiting at the edges of BB plywood where climb cut of *just the last tiny bit* avoids unsightly blow-out.  Doing this safely requires appropriate technique to ensure control of the router is maintained.

    CNC is a totally different thing because the router and/or table are mechanically controlled, so climb cutting is usually preferred for CNC to avoid blow-out.

    The difference in circumference for a 12" hole is 2% (~1.6"), using a 1/4" bit, which is why I didn't adress this issue. Unless you're cutting a 4" hole with a 1" tool, it doesn't really matter 😉

    Surface finish is a different topic thou. There might be differences.

    For regular work I'd recommend a simple straight bit. Quarter inch works great for this kind of stuff, if your router can spin it up to 15-20k RPM. I simply used a very long screw as center axis for my circle jig, which also held the cut out piece to the board beneath. I clamped the actual work piece to the board with two bar clamps.

    A dust collection system is also nice to have!

  20. 9 hours ago, SME said:

    Hey!  You need a jig to cut circles if you don't have one.  Parts Express sells one, or you might be able to buy one from a woodwork store near you too.

    Definitely learn as much as you can before you start trying to route.  It's important to clamp the work tight.  Also always route in the direction that makes the spinning cutter bite the wood from behind.  When the spinning blade hits the wood, it deflects the router in the opposite direction, and it's always safer and more stable to be pushing against the deflection rather than trying to "ease it forward while keeping it from flying out of your hands".

    Keep us posted.

    When "cutting" with a (fixed) router (like you do when cutting out circles), it doesn't matter which direction you go, since you're removing material with both sides of the bit.

    When cutting along a guide rail or something, cut the way that the router pulls itself towards the guide (e.g. from left to right, when the guide is behind the router). You don't wanna climb cut unless you're using a CNC and you're removing less than say 60% of the tool diameter during a roughing pass. That's at least what I experienced so far.

  21. 5 hours ago, SME said:

    Is that a sealed alignment or something else?  It sure does drop like a rock below 20 Hz.

    It's a little hard to figure out what's going on. The longer I look at the picture, the more confused I get. What looks like horn folds seems to be bracing. What looks like vents (bottom corners), seems to be a dead end...


  22. 1 hour ago, klipsch said:

    Nice. Looking forward to seeing you and your build thread for those :)

    Maybe use the cnc to build some Danley Jericho-esque mains for your PA 😁

    Yea there will be a build thread. I hope I get the time to do those in the forseeable future.

    And no, I'm not interested in high powered point sources for PA. Line-Array or small scale it is for me. I am toying with the idea of trying myself on designing line array elements. Some B&C drivers just scream 'buy me'... The CNC will surely not be bored!


    @Tahoejmfc I would've imagined that burning wood with a laser would generate some nasty smells from the wood itself (not just the fumes from the cutting). Since you're not running Nitrogen it's actually burning and not just vaporizing, isn't it?

    The idea with the driver wasn't a very serious one, but it would certainly be cool if you could get some actual sound from the cab :)

    Maybe @Ricci can fit a RF19 inside the 1qft cab.

  • Create New...