Be careful what you do.
Yes, the resistor the coil is wound around should be non inductive. A 10ohm 2 watt carbon or metal film is OK.
Bypassing R7 ? which circuit are you looking at ? Mine and the redrawn one you linked to have R6 and R7 reversed. Adding a 6.8nf across either would be a disaster. Across the 1k and the gain at hf would rise dramatically, across the 33k and the amp would probably self destruct as it would cause instability. Are you sure it wasn't 6.8 pf (pico) across the 33k ? That might help with stability issues but shouldn't as a matter of course be needed.
Changing the 27pf's to 100pf is an unknown, I wouldn't like to say without trying it for real and seeing what the effect was on a scope stability wise. I would leave them as they are for now.
I'm a wimp, remember 
. I won't attempt anything that doesn't have a clear basis - at my level of understanding So, increasing the gate stopper resistors; replacing R15 in what I would call the Zobel network (but I think johnnyx referred to as a snubber) with the same value (4R7) but not wirewound, and taking 0v line back to the power supply from a point away from the input signal's 0v was it.
Your kind link to the earthing discussion was a good reminder - long ago I was fortunate enough to know a very skilled and practical audio engineer (though too young to realise just how fortunate) who gave great emphasis to a star layout. To give due credit to Dave Goodman the accompanying leaflet did recommend using a single earth point - even though the PCB subverts that by leading the builder to use the same pin for Power 0v and the input signal earth.
. I won't attempt anything that doesn't have a clear basis - at my level of understanding So, increasing the gate stopper resistors; replacing R15 in what I would call the Zobel network (but I think johnnyx referred to as a snubber) with the same value (4R7) but not wirewound, and taking 0v line back to the power supply from a point away from the input signal's 0v was it.Your kind link to the earthing discussion was a good reminder - long ago I was fortunate enough to know a very skilled and practical audio engineer (though too young to realise just how fortunate) who gave great emphasis to a star layout. To give due credit to Dave Goodman the accompanying leaflet did recommend using a single earth point - even though the PCB subverts that by leading the builder to use the same pin for Power 0v and the input signal earth.
That's very convenient... I made a mess of the first resistor I used to wind the coil around and in replacing it I have several 10R 1W left over. Two in parallel for each Zobel will give them a useful home.
Thank you for all your help - as well as getting me going you have given me back a little of the confidence I will need for what was supposed to be my initial project. I have build quite a variety of units but that was, and certainly seems, long ago.
I have asked for some basic electronics texts for Christmas, hoping to improve (and modernise) my theoretical understanding too. Meanwhile its back to cutting the Aluminium plate to house these two.
Your generosity (and that of a large number of skilled and knowledgeable others here) is much appreciated
Thank you for all your help - as well as getting me going you have given me back a little of the confidence I will need for what was supposed to be my initial project. I have build quite a variety of units but that was, and certainly seems, long ago.
I have asked for some basic electronics texts for Christmas, hoping to improve (and modernise) my theoretical understanding too. Meanwhile its back to cutting the Aluminium plate to house these two.
Your generosity (and that of a large number of skilled and knowledgeable others here) is much appreciated
I see there are a few of these ready built on ebay for sale.
Maplin Mosfet Amplifier boards, 100w 8 ohms, 150w 4 ohm | eBay
Maplin Mosfet Amplifier boards, 100w 8 ohms, 150w 4 ohm | eBay
Your welcome
The 10 ohms in parallel will be fine. This lateral FET amp will give many of todays amps a run for their money, and subjectively quite possibly better a lot too.
The electrolytics will probably be losing capacitance by now.
I built one in the 1980's for mobile disco use but it wasn't loud enough.
Instead I built the Maplin 225WRMS 2n3055/mj2955 based amplifier and that was seriously loud.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Measuring cap values isn't always a definitive test. In the same way a battery can still read OK voltage wise even though its discharged, so caps can read OK value wise when in fact they are no good. What happens with caps is that the E.S.R. (equivalent series resistance) rises as they age and dry out. So you can have a 47uf cap that reads OK but its a bit as if there were a high value resistor in series with it.
Gentlemen,
With no intentions of threadjacking, I would like to forward a different
question about this amplifier.
Reading about this amplifier over the internet, it was emphasised to have
high beta and low capacitance transistors in the second stage driving the
mosfets. Unfortunately, the original transistors specified in this position
are unobtainium. My question is, is it possible to modify the circuit such
as to use Sziklai pairs in the second stage ? Would I be altering too many
parameters by doing so?
Best Regards
Selim
With no intentions of threadjacking, I would like to forward a different
question about this amplifier.
Reading about this amplifier over the internet, it was emphasised to have
high beta and low capacitance transistors in the second stage driving the
mosfets. Unfortunately, the original transistors specified in this position
are unobtainium. My question is, is it possible to modify the circuit such
as to use Sziklai pairs in the second stage ? Would I be altering too many
parameters by doing so?
Best Regards
Selim
Using Sziklai pairs would certainly change the whole operation and stability margins of the amp needing a redesign.
I doubt the original design is that critical on driver type and there will be modern replacements. The 2SC2910 comes to mind. Also Nikko Electronics list the Hitachi originals as available at £1.09 each.
I doubt the original design is that critical on driver type and there will be modern replacements. The 2SC2910 comes to mind. Also Nikko Electronics list the Hitachi originals as available at £1.09 each.
Oddly enough, when I built my Maplin MOSFET amp, I started with a bare board and bought the transistors from Nikko/Dalbani. To give you an idea of how long ago it was, they even had the original Hitachi output devices. :O
I also remember the trepidation of switching it on, as it was the first power amp I'd ever built. Luckily it worked fine first time.
I seem to remember the driver transistors being some quite beefy parts in extra-tall TO92 packages. Something like the KSC3503 would probably be a good substitute if you can get it.
I also remember the trepidation of switching it on, as it was the first power amp I'd ever built. Luckily it worked fine first time.
I seem to remember the driver transistors being some quite beefy parts in extra-tall TO92 packages. Something like the KSC3503 would probably be a good substitute if you can get it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.