peniku8

I built a loadbank to bench-test audio amplifiers on a resistive load. The loadbank consists of 16 3KW 220V heating elements, which allows me to test any combination of 8 8Ohm clusters for up to 4 channels. More on the loadbank in this post. Results: Sanway FP-13000 (MKI) t.amp TSA-1400 t.amp TSA4-300 Testing methodology: Calibrating the signal chain without an amp in the loop Getting a frequency response measurement of the amp Determining the amp's maximum unclipped output voltage (Vmax@1khz) with no load connected Adjusting my audio interface's input gain to scale the input signal at Vmax to -12dbFS (optional, just for convenience and consistency) Doing a reference recording at Vmax@1khz (any other known uncompressed output voltage would work as well) with no load connected Feeding the amp various sine waves and recording the maximum signal levels generated into various loads If clipping (>0.5%THD) was observed during step 6: Scale back the signal level in 0.1db steps until the THD drops below 0.5% Calculating the power into a given load with the level difference between the calibration recording at Vmax and the under-load recording Vmax' To 3: I used REW to do this. I sent the amp a 1khz sine wave and monitored the signal return with REW's RTA. I increased the level until THD exceeded 0.5%, then scaled it back by 0.1db. For amps with built-in soft limiters I approached the maximum possible outut voltage for the minimum possible input voltage in 0.1db steps. The generator in REW fades in, which results in 3.5 cycles (@1khz) being between -6db and 0db. That is equivalent in power to 1.75 cycles at 0db, which is why I generally add 2ms to the duration charts for the results. Not that it really makes a difference, but hey, I'm doing it. To 5-7: I record all the signals with Cubase 10 Pro. Vmax is determined through the Pmillet's voltage meter readout. I compared it against my HoldPeak TrueRMS meter and both were always within 1%. You can see my test environment in Cubase below. I checked THD of the recordings with MathAudio's THD Meter and noticed that as soon as the sine wave started looking clipped, it approached 0.5% THD. A very convenient way to quickly see if I need to reduce the output for a certain load (I did so in 0.1db steps if needed). I check all results with the THD meter notheless. To 8: I export the recorded files and open them in Adobe Audition. Audition can display peak levels under the cursor, which makes my life much easier. I use Microsoft Excel to do the calculations for me and plot the charts with. I feed it the reference dbFS(RMS) value and its corresponding RMS voltage, the recorded db value and the load it was connected to. Excel calculates the generated power and plots it over the user defined time axis. This is what my excel table looks like, with the Sanway FP-13000 into 1x4Ohm and 1x2Ohm as an example: Everything in green is user input for easy navigation. Power numbers posted will be the 100ms value. If any value within the 10s window deviates from that value by more than 5%, the power over time graph will be posted as well. Default time stamps are used when the changes are linear (left example above). I will adjust the time stamps when the default time stamps wouldn't represent the actual power over time plot accurately (right example above). The whole setup looks like below. The supply line is 6mm² (9AWG) 3x230V 32A. I log the voltage and current readouts of my power distributor. Speakon cables are 4mm² (11AWG). Even after several back-to-back full duration tests (2 minutes each) with ~2KW dumped into the loadbank, it was still completely cold to the touch from the outside. The tank holds about 15l of water.

Got an impedance sweep outside today. And gave them some music. They have lots of headroom for a cab this size for sure.

I just saw that the Eminence driver is also available in Europe now. I'd get three 21DS115 for its price here and even the 21IPAL is still cheaper 😐

Observations: Actual Frequency response vs Simulated HR+Edge Output: I didn't expect the native frequency response to be +- 1db from 35 to 100Hz, that is very nice. Average sensitivity is 89db at 2.83V @ 2m groundplane outside from 35-100Hz. The HR response here has been scaled down by 4.5db, the measurement was 1m, normalized to 2m (-6db). I don't know how the sensitivity ended up 1.5db higher than the sim, but I ain't complaining. Feeding the cab a 28.3V sweep showed zero compression between 40 and 80Hz. I didn't get a louder sweep yet, but the cab started to produce wind noise with 70V signals between 30 and 50Hz. It was hard for me to make out a tuning point as there wasn't a clear excursion minimum, but it seems to be 37-38Hz. I will get an impedance sweep tomorrow. Conclusion: I'm not entirely sure if this is actually the case, but it seems like the shaped port made the peak from the port loading a bit broader, which is why the peak is much smaller compared to the sim and the sensitivity around the peak is a bit higher. Maybe this is also the reason why I had trouble finding the excursion minimum. The tuning is quite low for a 12" 85L (3cuft) PA cab. The back chamber is about 50L (2cuft). I might go for a slightly higher tuning in v2, making the port entrance flare a little bigger and maybe adding some port area. The average port area is 120cm². It wouldn't hurt (ergonomically) if the cab was like 1cm wider. F6 is 31Hz, which is a good 10-15Hz lower than most PA subs this size. And most importantly: I like the sound!

Colour me impressed, whatever I throw at it, I can't get the ports to make excessive noise. Tuning came in at around 38Hz and even 35Hz sines at Xmax produce very tolerable wind noise. The results are very impressive and they also sound great without a dsp dialed in yet. I designed them to have increased sensitivity where excursion is low (slightly rising response towards the top end plus a bump around tuning), which will be corrected via dsp, since those subs were never intended to be used on non-dsp amps. I will get a free field measurement tomorrow. Unless it rains. They weigh just 19kg (42lbs or so) and have an external volume of 85L. I designed them to realize smaller shows alone (weddings etc.). They're also surprisingly stable when I pole-mount my EV ZLX12 on top. This is a great compact setup. I was going to do like 5 weddings with em next month, but the current health crisis delayed those by a few months. No actual gigs in the near future 😞

Finally! Measurements coming soon. If the weather is on my side.

That is a fairly high Qts driver, which is not ideal for "small" cabs like this. The sims don't look good either. Voltage sensitivity is really rough and hornresp says it has 5db less headroom than a 21DS115.

Blame those rough raw gigs back in the days..! At some point I started wearing ear protection for rehearsing and don't even know how I could endure playing the drums without em pretty quickly. I'm so happy that our band uses a full blown IEM system now, so I don't have to kill my ears with every gig. That nasty china cymbal...!

Yea I'll likely cross it at around 1.5k. I think a three way will be the better choice if your crossover will be that low. You'll need to get rid of all reflections which make its way into the mic. Since the mic is omni, you basically have to eliminate all reflections ideally (resulting in an anechoic chamber). Your horn is quite big so the mic distance should ideally be at least twice the largest dimension of your baffle (the diagonal in most cases which results in a mic distance of over 2m for you), but the closer your mic is to your sound source, the louder it is in relation to the reflections. I think you'll get decent results if your mic is between .5 and 1 meter away from the speaker and you hang blankets on the sides of the speaker and the mic. Carpet on the floor and another blanket ceiling should do the trick. The blankets should not be touching the speaker or they will mess with the edge diffraction. And EQ the drivers after the crossover if that's possible with your hardware. My next DIY project will be driven by a Hypex FA123 per speaker which has more than enough dsp for the task!

Cool topic, I wonder why I haven't discovered it yet. For speaker measurements I'd recommend doing those in a anechoic room (outside for us plebs). I think speakers with a native flat response over ~500Hz sound better than speakers which have been EQ'd flat indoors. Our brain basically filters out reflections to a degree, which the measurement mic does not. Even if you end up measuring a spike and EQ'ing that away, you might percieve it as lacking in that region then. If you have enough amp channels/dsp channels you could go with the B&C DCX464 as alternative to the BMS. It will have higher dynamic range (111db sensitivity+almost 400W power handling?! dear god...) and probably have a wider operating range due to the dedicated mid. It's priced similarly. I'm planning on doing a build with those and possibly something like a 12NDL88 or some 10" drivers. I'll also play around with the DE360. It won't go ultra loud but the HF response is super smooth with the new diaphragm material. Plus that one is only like 60€! B&C have optimized phase plugs inside their CD's to help with that issue and if you built some into your horn throat it should help. Would probably be too much effort for a single DIY project thou and I have no clue what those would look like.

Nice joints there! 😊

That's what I have an emergency stock for..!

I staple gunned foam to the sides to reduce distortion, especially out of band distortion when running at high output levels. Any sort of filling will reduce maximum output capabilities. PA subs usually don't use filling for that reason. If you have a very nasty resonance resulting in a bad waterfall plot, using some (lots of) filling will enhance SQ, but reduce sensitivity and efficiency. I'd line the sides with foam and only add some actual filling if I notice anything problematic when measuring/listening.

It was +75% of the shorter dimension in my case, since the port opens to the side of the longer dimension in my cab. It's a guess, but the tuning of yours will probably come in between +50% and +75%. What are you trying to achieve with the filling?

It will drop the tuning, but probably not by a noticeable amount. The cab will still be like 3cm off the ground and there are no side walls in addition to that. Also, the air pressures outside the cab and inside the cab differ, which probably also has an affect on how the geometry behaves, which is probably also insignificantly small. With my cab, I got the tuning spot on when I added 75% of the vent width to the length, with a vertical port. But that‘s only due to the weird port entry geometry and the fact that the port is only about two port widths from the rear wall. And the handles... Your port is pretty long to begin with, so the difference will be like 1Hz or so. I‘d not care much.

Keep in mind that the video above is one channel of the FP-13000 running full tilt into the impedance minimum of the 1/3 vent SKHorn there, which basically feeds it like 5KW since the two 21DS115-8 are in parallel. I never run into audible port chuffing during regular use, since I roll off the below 20Hz stuff usually.

Gotcha. No I haven't done anything custom with extruded aluminum profiles yet. All aluminum components you see came with the machines and are not off the shelf profiles. They've been sturdy enough for me to never break anything. All moving parts on the table saw are aluminum and everything else is steel. The machine is over 2000lbs, but using it feels pretty easy. I thought about doing the CNC stand with Bosch/Item profiles but my dad offered to make one from wood, which is what you see there. I wish I had welded a steel frame. The wooden construction moves too much for my taste.

I have not come across that feature yet. So you'd say it's reliable? Did it predict the actual room response/dispersion well enough?

If you reverse the butterfly layout on both subs, you could join the two volume boxes to a simple case. Brace it a little and add some PE foam and you have a mixer case or case for mic stands w/e you're carrying around with you

What exactly are you referring to? The T-Slot table?

Well if he is using the cab at home it's probably the room. I have a null at exactly 63Hz in my room, so I also feel like the SKHorn is lacking something. When there are nulls, there are peaks somewhere else. If I stand in the 63Hz peak it feels like I'm in a fist fight with Mike Tyson.

I'm unsure if the drawing is accurately to scale, but 90cm of port length seems close enough if you add half of the port height. This is a very large port. I like to have at least 100cm² per litre of displacement. For the 21DS115 that is 5l and I ended up using ~550cm² in a design that prioritized space efficiency. Low port compression is nice and everything, but I'd try to see if more back chamber volume isn't more useful in this case. I also think a meter of port length is quite a lot. A port this long on a 6th order bandpass would have me keep a close eye on group delay for sure. The SKHorn essentially uses less than half as much port area with 50% more capable drivers and the compression results are very tolerable until output exceeds 130db from 30Hz and up. Cover the top part of the cab in Warnex inside out so you can use it as transport box for something else 😊

Designing this thing on paper is pretty hard, I'm sure. Whenever I have a cab idea, I'm putting the whole thing together in Inventor and grab the measurements off the actual model, which I then input into HR. It's a lot of work for a concept which might be really bad, but it leaves very little room for error. This way you can create a very accurate feedback loop to get to the desired result. What's the actual frequency response of the cab? Did you find semi inductance parameters somewhere?

If you don‘t mind the short drive through Luxembourg, I could get you the 21DS115 for around 450 a piece or cheaper depending on how many you need. Just in case you want to get more nice drivers some time in the future S1 doesn‘t have a huge impact on the models. After all, S1 itself is not even in the path between the driver and the horn mouth.