Maplin MosFET Amplifier Ga28f construction thread

Seems many cut their teeth on these modules in the 1980's as I did.

Built my 1st stereo amp with them along with RIAA preamp etc and it sounded great.

Then while 'testing' at my lodgings, the police turn up and said there had been a complaint. I told them ' why didn't next door just ask me to turn it down?' and they replied ' the complaint was from somebody in the next street' .... still makes me smile to this day :)

PS the Celestions i was using at the time didn't last long - they shuffled off in a puff of smoke lol
 
Bespoke Constructor
Joined 2006
Paid Member
Hello happy forum members ... I have a soft spot for this amp as this is how I started out in electronics going to my local Maplin store . then made this lovely mosfet amp and nearly shook my parents house to bits, just love the 80's when I started making these amps
so now that I have my new PCB have decided to make a test bench to make sure all is ok with the amp before I commit to a pair of expensive cases etc
 
Last edited:
Bespoke Constructor
Joined 2006
Paid Member
so some pictures ... the mosfet amplifier mounted on the test bench chassis
 

Attachments

  • IMG_0497.jpg
    IMG_0497.jpg
    449.2 KB · Views: 331
Bespoke Constructor
Joined 2006
Paid Member
few more pics .. I have used a Velleman speaker protection module but have made a few mods
Like one heavy duty relay, and modded to PCB so both input channels can be used for a mono amplifier (dual redundancy)
plus played around with the timing capacitor for power up start up etc
 

Attachments

  • IMG_0501.jpg
    IMG_0501.jpg
    575.5 KB · Views: 110
  • IMG_0500.jpg
    IMG_0500.jpg
    416.1 KB · Views: 88
  • IMG_0499.jpg
    IMG_0499.jpg
    451.8 KB · Views: 102
The 100W Maplin amp was a great implementation of the original Hitachi design and I've put together loads of them. The same circuit works well with HEXFETs too with minimal circuit changes. I cant tell the difference between this design and more advanced efforts.

A great trip down memory lane really!
 
I built two Maplin mosfet amplifiers back in 1993. I noticed a difference in the circuit regarding the 10 ohm resistor in parallel with the output inductor. The instruction leaflet stated that the resistor is not provided or required.
Older Maplin mosfet kits included the resistor and stated that it was for damping the inductor to prevent ringing.
Has anybody on here done a square wave test on the amplifier to determine if the resistor is required or not.
 
Hi Guys,

Before starting a build of the famous Maplin 100W Mosfet amp, I have been experimenting with LTspice (original file by Mooly?).
I have 2SA872A trannies for the LTP and Exicon 10P20/10N20 latfets for the outputs.
I cannot find the original 2SB716/2SD756 but have plenty of KSC992/1845 instead which look pretty similar (120V, high hFE, low Cob).
Some have used MPSA42/92 but the ones in my stash have pretty low hFE <100.
The simulation is very promising with a 0.0005% THD @1W.

I have tried next to add a J112 CCS and a current mirror, keeping the LTP/VAS/output current values nominal.
In these conditions the predicted THD is 0.0025%.
Is it expected to have a higher THD with CCS and CM? Do I miss something?

Jacques
 

Attachments

  • 10_Maplin_CCS_CM.asc
    7.6 KB · Views: 76
  • 09_Maplin.asc
    7.5 KB · Views: 75
  • Maplin CCS CM.PNG
    Maplin CCS CM.PNG
    81.7 KB · Views: 511
  • Maplin.PNG
    Maplin.PNG
    77.7 KB · Views: 516
Bespoke Constructor
Joined 2006
Paid Member
Hello

A couple of pics of the final PCB completed .. I have used 90 degree angle pcb tags for all connections and a phono connector for the input.
I know lots of people complain about hum with this PCB design but in my configuration I have always had very good results ??
 

Attachments

  • IMG_0559.jpg
    IMG_0559.jpg
    340.7 KB · Views: 354
  • IMG_0560.jpg
    IMG_0560.jpg
    464.5 KB · Views: 340