Announcement

Collapse
No announcement yet.

Tent Sale, new project to start

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Tent Sale, new project to start

    So after a very brief trip into the back of the room at MWAF, I was off to PE for the tent sale. I have an old design compact combo amp I designed and built back in college that has been out of commission for a long time. Here are the problems:
    • Power amplifier of my own design, was never super stable and with the +-25VDC power available only managed about 25W into 8 ohms, and 40 into 4. The amp died some time ago and I've never revived it
    • I didn't go for loudness with the driver choices, had very little money for drivers, and knew very little about crossover design. The woofer is the MCM 8" poly cone driver and I used a Radio Shack direct radiating piezo tweeter (really, really cheap)
    The box is about 1 cube, less electronics inside and has hefty power supply available for an amp in a 140VA transformer and 40,000uF of filter. The preamp is my own design and works well, as does the power supply and a delay turn on speaker relay. The back panel carries the amp with a heat sink to the outside. I decided a new amplifier module, and a driver update was in order. A bridged output amp would quadruple my power output with that voltage, and I decided to give up a little low frequency extension for some sensitivity. Here's what I went with:That's about $100 at the sale so far, still need crossover components though. With the 4X power and +6dB sensitivity of the drivers I'm looking at 12dB increase in output from this sucker. The amp is intended for a portable bass guitar solution, but I like full range sound with my bass and intend to design this sucker with a reasonably flat frequency response. The PA200 has a hefty break up at 2K, and that peerless tweeter may not be awesome, so I don't expect super fidelity but it should sound pretty good as a mono music box when needed. So far, the woofer models with extension down to 50Hz in the cabinet after shortening the port tube to an inch or so.

    The fun part will be updating the electronics, then making actual measurements for a crossover design.
    Click image for larger version

Name:	PPA_newparts.jpg
Views:	3
Size:	65.0 KB
ID:	1289351



    Click image for larger version

Name:	PPA_ampguts.jpg
Views:	2
Size:	75.6 KB
ID:	1289352
    Attached Files
    Electronics engineer, woofer enthusiast, and musician.
    Wogg Music

  • #2
    Nice project! I remember looking at those tweeters in the Radio Shack catalog, I think they were $5 at the time.

    Comment


    • #3
      Thanks Chris!

      ​So I had a minute to put together some virtual ZMA/FRD files and model, holy monkeys that tweeter is hot. Here's the story on that selection... I had a Morel MDT-37 in my basement somewhere that would have been sweet, except Saturday morning I couldn't find it. Tore the whole basement up looking and can't remember where the heck I put that thing, so right before I left for Dayton I scoured PE for a horn loaded dome that would do the job. In my haste, I misread the sensitivity on that tweeter, thought it was low 90's instead of 99dB/W. Whoops... now it looks like I'll have to pad the crap out of it to the tune of 30 ohms-ish to bring it in line. I've played with PCD a bit, when I go for reasonably flat I lose phase tracking while locking phase results in poor flatness. Regardless, this is theoretical stuff so far, so the real measurements may be quite different. We'll see. Here's a shot of a quick sim check (not perfect of course).

      Click image for larger version

Name:	sim1.PNG
Views:	3
Size:	71.8 KB
ID:	1289809


      Attached Files
      Electronics engineer, woofer enthusiast, and musician.
      Wogg Music

      Comment


      • #4
        Did you try a zobel or some resistance across the tweeter terminals in combination with the padding resistor? 30 ohms sure is a lot. I was expecting you to say you found the Morel as soon as you got home

        Comment


        • #5
          Nope, only worked on the network for about 10 minutes last night to see about where it would land. A parallel resistor is a great idea, that'll help reduce the required values a bit. I have to keep playing with it, and also fix the box model mismatch in response modeler (see the step at 300Hz?). Regardless, this part is more of an academic exercise, I need to get an ARTA / LIMP measurement rig up to get the real thing after I update the cut-out for the tweeter, get the amp wired up, and re-finish the box with duratex.
          Electronics engineer, woofer enthusiast, and musician.
          Wogg Music

          Comment


          • #6
            Here we go, much better with a parallel resistor on the tweeter and some minor value tweaks. I'm now within +-3dB territory and phase tracking is good. The impedance runs just under 5 ohms in the midrange, the amp module should handle that fine but my power supply may sag a bit. BOM of about $26 in components. The Morel tweeter will show up immediately upon putting this together

            Click image for larger version

Name:	System Response.png
Views:	3
Size:	79.2 KB
ID:	1289894 Click image for larger version

Name:	System Impedance.png
Views:	2
Size:	63.2 KB
ID:	1289895
            Attached Files
            Electronics engineer, woofer enthusiast, and musician.
            Wogg Music

            Comment


            • #7
              Originally posted by wogg View Post
              So after a very brief trip into the back of the room at MWAF, I was off to PE for the tent sale. I have an old design compact combo amp I designed and built back in college that has been out of commission for a long time. Here are the problems:
              • I didn't go for loudness with the driver choices, had very little money for drivers, and knew very little about crossover design. The woofer is the MCM 8" poly cone driver and I used a Radio Shack direct radiating piezo tweeter (really, really cheap)
              Quite OT - I used a pair of those tweeters along with 4" Jensen car full-rangers as my main computer speakers for years (they are now surrounds in my basement). With a bit of graphic EQ, they didn't seem terrible at moderate volumes.

              Comment


              • #8
                Originally posted by chrisn View Post
                I was expecting you to say you found the Morel as soon as you got home
                He'll find it when he's done with the build .

                Comment


                • #9
                  Today's the day for some box re-finishing. The tweeter mounting hole was re-routed to fit the Peerless, and after some light sanding and minor wood fill touch up it'll be ready for Duratex.
                  Electronics engineer, woofer enthusiast, and musician.
                  Wogg Music

                  Comment


                  • #10
                    Design note on the bass alignment... while the PA200-8 looks OK in a 0.9 cube box tuned to 55Hz response wise, there's a slight problem with the power handling and low Xmax.

                    Click image for larger version

Name:	NA.PNG
Views:	1
Size:	36.0 KB
ID:	1292556


                    Click image for larger version

Name:	EP.PNG
Views:	1
Size:	34.3 KB
ID:	1292557


                    The input power graph shows I'll breach Xmax at only 35W around 80Hz. While this would be a huge problem for a HiFi speaker, I'm not worried about this for bass guitar. The speaker has a much larger than 3mm travel before Xmech hits, so I'm in no danger of damage, which means what will happen is increased distortion. For a bass guitar, that distortion in the low end is likely to end up pleasing, like a layer of warm grunt that may make the whole thing sound meatier than it is.

                    For extra expansion I plan on incorporating a switchable line out on the back that will toggle between a full range line out, to a 200Hz low pass output. When set to low pass, the internal amp will get a matching 200Hz high pass, cutting off that high distortion area and passing it to an external amp and speaker. That will reduce distortion, allow higher output and when coupled with a powered bass module this thing should support small gigs. I've been toying with the idea of a dual 10" woofer sub powered by a plate amp for the low end, dimensioned like a stand for the little amp. That will be sometime in the future and would be a sweet portable setup.

                    Here's the schematic for the crossover mod, 3rd order sallen-key filters. Values to be finalized, but board layout is done. Included in this is a 3rd order high pass at 40Hz (maybe 50) that will be the default for the internal amp and protect the speaker below Fb.
                    Click image for larger version

Name:	PPA40 Crossover Board Schematic.png
Views:	1
Size:	55.8 KB
ID:	1292558


                    Electronics engineer, woofer enthusiast, and musician.
                    Wogg Music

                    Comment


                    • #11
                      The Duratex turned out lovely and the speakers are in. Tonight's work.... passive crossover build! Done with the help of a little IPA of course. I did this without the old test lead check... was really careful with the connections and metered them out when it was done. After it was assembled, I ran the speaker straight from a function generator to check the hand off from woofer to tweeter, to the ear it sounds smooth over a sweep. The back panel is still drying after installing the new metal back plate, so I can't seal it up and test quite yet.

                      In other news... never underestimate the expansion of Gorilla Glue, I have a lot to clean up.
                      Electronics engineer, woofer enthusiast, and musician.
                      Wogg Music

                      Comment


                      • #12
                        Looks nice and organized, especially the crossover.

                        Comment


                        • #13
                          Thanks Chris!

                          Build updates for the weekend:
                          I had to re-do the back panel seal due to the gorilla glue mess, but it's in there now. The 120V mains were re-wired, I didn't like the wire I used back during the original build. I plugged it in and tested that sucker out, and it didn't explode. The power supply and pre-amp are good, the speaker relay was bypassed forever ago, and I couldn't remember why. It turns out there really isn't a delay and the relay kicks nearly immediately. I'd have to pull that board and re-trace to fix, but it doesn't bother me much unless turn on thump is a problem. It does enable the 2 color LED on the front and disconnects the speaker before the filter discharges.

                          After some drill and tapping, I mounted the amp board to the heat sink and wired up the harness.

                          Click image for larger version

Name:	2016-07-31 14.18.26.jpg
Views:	1
Size:	344.4 KB
ID:	1293273

                          Unfortunately, immediately on power up the fuse blew. After carefully checking power supply connections and everything else, I couldn't get that power to sustain. The speaker pops, so there's definitely something up with the amp. In fact, the 1A fuses popped frequently even without the amp connected at all due to the inrush current charging that 40,000uF of capacitance. I'll have to find some slow blow versions for that. As for the amp, it may be suspect. I couldn't find any thing wrong with it or the mounting, no shorts or opens and the pre-amp was fine. I'll bring it into work this week to scope it out and make sure before I pursue an RMA.

                          I did seal it up and run some test leads out the port to check out the speaker. I ran an RTA, and did some single channel ARTA tests on the speaker alone, and it performed as expected. My measurement capabilities are a bit weak, so I didn't do save any graphs yet. Music sounded decently balanced, little light on the deep end though. Once I get the internal amp working in it I'll work on real measurements and figure out the whole gating thing.
                          Attached Files
                          Electronics engineer, woofer enthusiast, and musician.
                          Wogg Music

                          Comment


                          • #14
                            Lucky for me the amp board is fine. I just had it on a bench with a lab power supply, generator and scope. Spec wise they set it up with 38dB of gain, and the sucker is flat to near 100kHz. The power supply was only +-25V at 1A, so there wasn't enough juice to see what it could really do. It managed 6VRMS into a 6 ohm load before hitting that 1A current limit, which was enough to get the heat sink nice and warm.

                            Tonight I'll pick up some slow blow fuses and hopefully be in business.
                            Electronics engineer, woofer enthusiast, and musician.
                            Wogg Music

                            Comment


                            • #15
                              Last night I put in a slow blow fuse, and it powered right up. There was a turn on thump due to my delay relay not actually being delayed, but not too bad... and then there was buzz, lots of buzz. The power supply ripple rejection is poor on that amp board. I ignored that for now and sent it some signal, which worked like a champ.

                              First impressions, ignoring the power supply buzz:
                              This little sucker is loud! The cabinet has never had that much power in it, so I'm finding all kinds of rattle and noises from tiny holes in the heat sink and the open back 1/4" jacks. I have some caulking / weather stripping to do. Also, the pots are dirty from a long time in storage, I need to clean those suckers up.
                              Playing bass, it sounds great. I need that 50Hz high pass implemented, the woofer cone flops about with the palm on the strings but other than that, this thing will easily keep up with a acoustic gig, or a quiet drummer.
                              Playing music the midrange is a bit forward and the highs are not as crisp as I like. The shelving tone controls bring in the bass punch and some crispness, but with a flat signal the speaker is a little narrow. That tweeter has a very narrow dispersion and seems to die off axis quickly, I would not recommend that for a HiFi speaker at all, but for this application it'll do well. The missing Morel MDT-37 would have been better.

                              As for the buzz problem, the amp board has a overly complicated regulator circuit for the op-amp pre section. They did that to accommodate a wide range of input voltages, but it appears the "regulation" is weak. I verified this by pulling the op-amp from the socket, leaving the chip amps powered up and active... no buzz. I swapped in a TL082 op-amp just to see if the NE5532 was more sensitive to power supply noise, no change in buzz but the power on thump went away so I may stick with the cheaper part in the socket. I've requested schematics from Juan-Jing, but am not holding my breath. My plan at this point is to bypass the amp's regulator circuit and use the +- 15V from the crossover board to power the op-amp. When you know the source power supply, a good zener regulator only needs 3 components instead of the 11 Juan-Jing put on the board.

                              Once I have the buzz sorted I'll do the ARTA measurements with it under its own power.
                              Electronics engineer, woofer enthusiast, and musician.
                              Wogg Music

                              Comment

                              Working...
                              X