Sunday, April 15, 2018

Bluetooth Speaker


2018APR15
   Began making a "bluetooth speaker". Using a BT receiver I found for sale on Groupon. It has RCA outputs and an 1/8 inch stereo jack for audio out as well as USB-B micro and USB-A jacks, too. The top wasn't difficult to remove, though a couple of screws are hidden. It doesn't have a battery, unlike the very small units you may find. From the description I thought it had a battery, being kind of heavy that way, but it has a weight placed inside without which it's very light. It worked well with the Grundig.
   Wired out the BT pairing (red) and status (blue) LEDs as well as the main pushbutton, which serves as power and as a means of initializing pairing and re-establishing a BT connection, for relocation on the speaker's front panel.
   I'm going to build an amplifier for it using the LM4871 (*2) which is designed for 5V applications. Not certain whether I'll have a panel-mounted pot because I won't know how much actual gain it has, how much control I'll need, till I build it. I am satisfied that the concept is sound, based on my test using my Archer amplifier.

2018APR21
   Built up one experimental channel of the LM4871 amp on the old breadboard.


  Decent volume with 20kΩ R_fdbk, but better with 33kΩ. No "0.39µF" for that input cap, but there's a range given for that (0µ1 to 0µ47) - so I used two 0µ1 in parallel [0µ2].
  I didn't bother with the 'Shutdown' pull up, just tying pin 1 to GND.


2018APR29
   Knocked out the speaker panel. Soft maple is a lot harder than soft pine. I don't have a scrolling saw or anything like that, and my single-speed jigsaw is only for super-hack, so I drilled holes around the speaker hole borders. Plan A was to remove the rest with a chisel, but chiseling with the grain wasn't working out so great and across it may as well have been rock (too much work). Plan B was rasping out the rest to fit. Cannot complete the job till the grille fabric, held up in post, is delivered. (It was "mis-sent", that's official USPS lingo, to another post office and has languished several days there already.)
   There was a little chip-out, but a bit of grille cloth can hide a multitude of sins - assuming it ever gets here.


2018MAY11
   Went with a sort of cross between "dead bug" and "manhattan" methods by cutting out 'islands' (instead of gluing down isolated pieces).
   I didn't order any components, drawing strictly from accumulated inventory.


2018MAY13
   Amplifier test 100% success, no "issues".
   Continuing with the speaker panel, I glued some grille fabric but it has some sag; not terrible but anyway. Using some 'modern' synthetic fabric, black, which needs to be stretched out, somehow, but I couldn't think of a way to do that. Maybe there's a method on the web somewhere ("instructables"?).
   Still haven't tested it all powered from the intended, 2-outlet USB "cube," - been running the BT-RX from that, with the amp from a separate linear power supply.
   I found some brass fender washers to use as bezels around the front panel knob holes. I found some inserts coarse threaded external with machine thread internal, but decided to fix the speakers to the panel just using wood screws. The 'inserts' were a total pain in the hole.

2018MAY20
   Completed the mounting of the speaker panel.
 

   Cut, drilled and screwed in two struts. Gluing the struts would have been easier but would likely have made their removal destructive. I haven't altered the cabinet, I've really wanted to avoid anything like that - though I had to bore ever so slightly the 4 existing holes for the dial bezel for the LED/LCD bezel. I thought that I had 8 screws, but I only had 6, so I used hot-melt on the brackets between the struts and the panel.
  [The seller wanted the cabinet sold or "it's going to Kutztown".]


   Full USB Power success, both with the "cube" and the 5V rechargeable. I made a USB_A cable for Amp Power that plugs into the USB_A receptacle on the BT-RX ─ so only one full USB cable is needed (a USB_A to USB_Bmicro). Having the 5V rechargeable option makes it completely "wireless".
   The sound is great, even better than I had anticipated.
   I need to place the BT-RX and the amp board on a panel and commit to some finalized wiring. Everything is loose and alligator clips, rigged temporary.


  Remaining on the to-do list are:
  •  routing and mounting the Status LED and pushbutton to the cabinet front panel
  •  building and testing the LED panel (threshold detector) circuit [Built, tested, but undocumented ─ as yet]
2018JUN13
  Of note - the cabinet is a Silvertone Model 1955 (c. 1936). Silvertone was the Sears store brand for their radios and, later, TVs and other home entertainment electronics. Here is a picture I found of one that's survived. They're 18 inches tall (2,160 cu. in.)
 [When I bought it, there was only the cabinet for sale, the seller had no chassis for it.]

2018JUN14
  Here's the first version of the threshold detector ─
  The inputs are connected to the amplifier at the junction of Ri and Ci (see the amp schematic from 2018APR21). They're pre-biased at Vsupply / 2 (or 2.5V). The R1 pot is the threshold trip adjust. I will post a demo, "soon".
  > > Have to add that that design is based on a circuit used in a Sharp/Optonica cassette that I once owned (a "peak" LED that went along with the VU meters). Really lazy "design", needs more work.
  And I'm not liking how it's putting the first transistor circuit in the Amp/LM4871 feedback loop either.
  I'll keep the diode-Or'ing on the input, to an op-amp buffer and then a comparator, the output of which will feed into a cap/hold-up circuit (better than) above.

2018JUN24
  Here's the improved circuit. It has a nice fade-out, less blinky.
I added the 1M because it was susceptible to stray 60Hz otherwise.  U1a is a unity gain buffer, U1b is a comparator, and the 10k pot sets the threshold.


2018JUL23
  Yesterday was a long build session, but worth it.  "Phase 1" is now Complete.

  Here is the threshold detector / LED driver [TDLD] board --


  Here is everything placed in the cabinet --
  Yes, there's a lot of gaffer tape in place (for now).  With gaffer tape you can hold down wires, and other stuff, without a firm commitment.  The TDLD is there on the right where you can also see a boost converter (5 to 12V).  I did the LED panel research with 12V and I wanted enough headroom for the op-amp, ergo implementing the boost mod (less than $5).

  Here's the front --
  I don't like the pushbutton (lower right), it doesn't look like it belongs. Its saving grace is requiring little pushing force.  I dug up a much smaller pushbutton to replace that with.  For now, pending work on "Phase 2", I'm using some hole-plugs for the holes on the left.
  The speaker panel can be seen there despite the grille cloth, but only when seen through the camera.
  Tomorrow I will upload a demo ("do not leave town!").
  [Special Thanks to Dennis Zhang for lasering the 'dial plate'.]

2018JUL24




Monday, January 15, 2018

Grundig - RTV 320 U


2018JAN15

  Picked up this tuner-amp (c. 1968). I was drawn to its having a shortwave band. Note that the high frequencies are on the left.
  Blew it out, cleaned it up a little - it's basically functional.
  The previous owner tried rigging in a couple of RCA jacks to bypass the euro speaker connectors and it had no dial lights.
  The back panel serves as in "interlock" of sorts. I removed the AC cord, drilling out the rivets. I may take a shot at fabricating a replacement back panel (from fiberboard or G10).
  I found an ebay seller with the euro connectors and another with a DIN-to-RCA cable. The Phono and Tape inputs are via DIN jacks.
  I replaced the dial lights. I must have been the first person to have done that. That's probably because doing so requires pulling the knobs and removing, extricating, the dial glass (a total PITA for dial lights). The lamp bases were practically frozen in their sockets. Curiously, they run from their own transformer secondary.
  The left and right audio power amps are each powered from separate secondaries of their own. Those Grundig designers were generous with transformer windings.
  The AM and SW switches have an issue, I have to tap them to get a circuit engaged. I should go back in, those contacts may benefit from some DeoxIT in/on the contacts. [see Follow Up]

  Tuning between FM stations results some noise like what you get from a noisy volume pot. I think this is the result of the receiver's lack of 'FM Muting'.
  The Volume pot is not noisy - which is surprising. Likewise, neither Treble nor Bass pot exhibit any noise and Balance is clean, too.

"Aus Bayern, mit Liebe"

  Follow-Up: Took the chassis out and sprayed some DeoxIT into the AM and SW select switches and it seems to good effect. They're sort of open at the back: spray and work 'em, wait, work 'em again, and repeat.
  The back panel calls out "7V" lamps and, given my 6.3V #40's, I was bugged about the line measuring 6.8v(ac) - so I put a couple of diodes back-to-back in series, knocking that down to '6v'. [The two dial lamps and the one behind the tuning meter are in parallel.] As their brightness was excessive, garish, reducing v_lamp "to spec" this way made that nicer.
  [Hadn't looked at the circuit board first time around, but upon going back in I noticed that the screws securing it to the chassis were soldered over.]
  I was considering putting this in the front room as my 'daily driver', but it doesn't have a tape monitor loop.
  Tuning SW stations requires extra patience because it doesn't have a fine tuning control. Nonetheless, I'd have been especially happy to have had this during the international broadcasting era. Sadly, there's not much to tune in anymore.

  11FEB2018 - Good results with a couple of Bluetooth receiver configurations, connected with DIN to RCA adapter found on (where else?) ebay.

Here's a hotlink to the schematic PDF.