Advertisement



  1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

fixing a Rel Strata III subwoofer amp

Discussion in 'd.i.y.' started by geoturbo, Jul 29, 2020.

  1. Arkless Electronics

    Arkless Electronics Trade: Amp design and repairs.

    An 8R driver will measure about 5.5R ish for DC resistance. 8R is very much an arbitrary figure.

    One pair of Hitachi mosfets can work OK with 4R but it much depends on the required power output. They are resistively quite lossy with about 1R Rds on and so best used in multiple parallel pairs... which is very expensive!
    Probably the most reliable output devices used in audio though;)
     
  2. davidsrsb

    davidsrsb pfm Member

    True, but a conventional 8 Ohm speaker has significant series resistance from the crossover inductor and the coil inductance comes in by 1 KHz.
    The sub-woofer is direct connected, so the high Rds and Vgs means lots of heat
     
  3. geoturbo

    geoturbo pfm Member

    Time flies and no updates.. thinking I should give an update here despite little time available..

    I am still not there but am hard headed.. I (think I) found a pretty similar schematic to the Rel one thanks to @davidsr suggestion:

    [​IMG]

    I added some heatsinks on driver transistors, routed R6 resistor to another exposed point due to the damaged pcb and substituted the original non working LD1 red led (which is in place of D1 and D2 in the above schematic):

    [​IMG]

    with a green led I had laying around:

    [​IMG]

    With the not working red led the dc at the output was constantly increasing till (I guess) positive rail voltage (I guess because I did not want to damage something else and turned it off earlier than damaging something else, I suppose..)..
    With the green led this dc was bouncing up and down in a more controlled manner but still the cone was going one way outwards.. heating up and not returning back..
    Now put a red led back in and now the dc on the output still bounces up and down, but the cone remains in center position and big uneven thumps are moving the cone..

    Not sure this is the right way to go but now it appears as a false contact might be there (I will doublecheck).. or maybe even a higher forward voltage diode should be used?

    I tried to address this question to Rel..

    or an inlet transistor might have failed together with the red led? let's see how this goes..
     
  4. davidsrsb

    davidsrsb pfm Member

    Try to use red, the Vf voltage is about 1.5V, which is the same order as two silicon diodes and just happens to give better temperature stability.
    Green is about 2V, which will make the LTP tail current too high
     
  5. geoturbo

    geoturbo pfm Member

    Thank you David.. I thought it was the other way around about forward voltage with colours.. I will measure the current one and should be having a couple 1n4148 around too..
     
  6. davidsrsb

    davidsrsb pfm Member

    Vf determines LED photon wavelngth and hence the colour.
    Using a red LED as a bias is a standard engineers toolbox trick for a stable current source
     
  7. geoturbo

    geoturbo pfm Member

    Actually I measured forward voltage of red led just slightly below 1,9V; I then put back green led and measured just above 1,9V.. put two 1n4148 diodes (voltage measurement is oscillating) but now the ac in the speaker remains in all cases..

    Thinking that I should find a 1,5v red led and see if things get fixed or (one of) the input transistors is not working properly.. my best guesstimate would be Q3 in the schematics (and Q3 on the board) which the one closer to the red led..
     
  8. Arkless Electronics

    Arkless Electronics Trade: Amp design and repairs.

    FWIW R14, 10K 1W in the schematic above is about the most common failure on the Hitachi topology.
     
  9. davidsrsb

    davidsrsb pfm Member

    1.9V on a red LED means either a very high current or you have a LED with a series resistor
     
  10. geoturbo

    geoturbo pfm Member

    Thank you, @Arkless Electronics
    I have substituted R14 on schematic (which is R8 on Rel PCB) with a 15KOhm 25W* Caddock thick film resistor (*I added a small heatsink to it, which should be fine for 5W or more, I measured little more than 100V of diff. potential on this resistor for a current of 7mA for little less than 1W continuous dissipation). Old resistor was cracked by heat, also the pcb has been pretty badly heated up around this resistor and the 4x driver transistors (mostly the two central ones to which I added small heatsinks too).

    @davidsrsb I gueess I will need to order a 1,5V red led then to understand whether the current is very high or if I have a led with series resistor.

    I repassed the soldering points with not much luck, pretty much sailing in the dark, not much fun so far, glad of your lighthouse support, thanks a lot guys.

    Thinking I could purchase some BC546 to substitute the three input transistors as per above schematic.. but not too sure these are the correct ones to be used; not sure it helps, but the three SMD transistors in place are having a "2D" marking on the upper part, I asked REL about these but reply is hard to get.
     
  11. geoturbo

    geoturbo pfm Member

    Ecce sound!

    Not one year has passed, and beyond all expectations with some help by a very valuable fellow forumer it seems the amp is working now.
    I will test it like it is for a while and pimping it is saved for later on.. whether torture tests will go well.. :)
     
    russel likes this.
  12. davidsrsb

    davidsrsb pfm Member

    Well done. By the way, "2D" is a MMBTA92, so a MPSA92 would be the TO-92 equivalent
     
  13. geoturbo

    geoturbo pfm Member

    Also worth to mention is the following link where a very close schematic to the actual thing is present down the thread:
    https://www.diyaudio.com/forums/sol...ors-bk-electronics-mf300-circuit-diagram.html

    Yep, along with another zillion of solderings (the board looks really awful now) I substituted the three "2-D" input transistors (I guessed/must have read they were MMBTA56) for through hole KSP92 TO-92.. hopefully they are close enough to your consideration @davidsrsb .
    I needed to try with different leds/bias resistors on Q3 which as far as I understood drives the hole circuit.. a bit of tinkering with these values was needed in order to put back output dc to an acceptable value smaller than 30mV or so keeping enough current flowing on Q3..

    The circuit seems stable with no load.. when I have time I need to fix a shorted chip (TL034 soic-14) on the crossover (ouch) and then if all goes well it will be testing time..
     
  14. DaddyDJ

    DaddyDJ pfm Member

    I am watching this with interest as I have a non working Rel Stadium III.
    I dont know if the amp boards are the same or even similar but I am sure I will get some insight from following the thread.
     
  15. geoturbo

    geoturbo pfm Member

    Have no crystal ball but there is an high chance that the schematic of your Stadium III is very similar to my Strata III as what is already indicated in the mf300 circuit diagram thread..
    my Strata III is specced for 100/150WRMS and has a couple of output trans, whilst your Stadium III may have two couples of output transistors (BUZ900+BUZ905) instead for more output power.
     

Share This Page





Advertisement


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice