18Watt.com

In my experience, the MOSFET needs a heatsink. Mine gets pretty hot.

Yeah it has a 2mm silicone heat transfer pad underneath the mosfet and the fibre board is 14mm thick.i will monitor it and see if there is a problem.taking some time out to learn the basics of using my oscilloscope now.

geoff 1965 wrote: ↑

Wed 06/16/21 12:05 pm

Yeah it has a 2mm silicone heat transfer pad underneath the mosfet and the fibre board is 14mm thick.i will monitor it and see if there is a problem.taking some time out to learn the basics of using my oscilloscope now.

The MOSFET will need to be mounted on a metal surface, with the silicone pad, to transfer its heat away, not the circuit board.

Okay will do,might need to alter the board to suit.i’ve also sourced some FQA9N90 mosfets which are a lot closer in spec to the NTE2377 on the schematic.

Why not place a smaller heat sink between the mosfet and the fibre board. For better heat transfer you could mount it without an isolation pad, if the heat sink is not connected with the chassis....

musicheals

Or mount the MOSFET vertically instead of horizontally, and then you have a lot of heatsink options without needing to worry much about the board.

Update on the little ecl86 plexi,i’ve made a small vvr board and now the mosfet is bolted to the chassis and is working good.
Also replaced that linear hi-fi output transformer with a Hammond 1750EP which is 15W and 8K5 primary and 8 ohm secondary “I can test now with my favourite eminence speaker” should be more suitable now with a guitar amp OT.
I bypassed the effect loop triode for now until i’ve finished tweaking the amp then I can concentrate on that last.

Fitted the tinted Perspex front and happy with it,also fitted a 5ohm 6amp thermistor just for the soft start protection for the mosfet,zener etc in the vvr section.
Decided to go passive series with the effects loop,it works for me and there is no resistors/capacitors involved,I did’nt want to go through the amp setting gain & tone then have to attenuate it after the buffer triode into the phase inverter.so I have just used the extra triode to parallel V1 and going to experiment with that.

VVR gremlins again!
when on full or minimum voltage the volume is fine,however if i sweep from full power down i'm getting massive noise,not sure if it's the mosfet but i changed back to the "hall's" layout except the screens tap which i have in my capacitor tester and has been no problem.the 32/32 can cap is brand new so i don't think filtering is the cause. here's my layout and also the halls version. good news is i'm getting some cracking tone/breakup with this plexi circuit and the ecl86's

Mosfet number 3! gone through all the components on the vvr board and fitted another new mosfet and all values are correct.the one thing coming to mind is the potentiometer,the volume was good on max and good on minimum with no noise,the problem was when sweeping up or down so maybe the carbon track was damaged when the first mosfet played up? this has me stumped but i’m determined to sort it out!
Also now the B+ has settled with all the tubes biased it’s a little too high,plate voltage diode rectified is 312V and 298V tube rectified,I don’t want to push the ecl’s over 300 so i’m going to drop the B+ 10 volts with a 5W zener.

2 little buddy’s have just arrived! one for the plexi ecl86 and one for the 18W amp.

geoff 1965 wrote: ↑

Fri 11/26/21 8:59 am

2 little buddy’s have just arrived! one for the plexi ecl86 and one for the 18W amp.
D4E10765-4CEC-4D88-B512-F0BD8936971A.jpeg
C654CA01-0301-4DDC-88CF-7C5CB75F16D9.jpeg

Niiiiiiice!!!

Bieworm wrote: ↑

Fri 11/26/21 12:44 pm

geoff 1965 wrote: ↑

Fri 11/26/21 8:59 am

2 little buddy’s have just arrived! one for the plexi ecl86 and one for the 18W amp.
D4E10765-4CEC-4D88-B512-F0BD8936971A.jpeg
C654CA01-0301-4DDC-88CF-7C5CB75F16D9.jpeg

Niiiiiiice!!!

Those two "Made in England" G12H-30s will sound big and solid! Hardly "little buddies"

Yeah ZP!
They were in a solid state carlsbro amp that was spares/repairs,date codes are October1971 and October1972,50 years old and look at the cones they are mint,these have been looked after.

Cathode capacitor open! that’s a new sprague atom as well the little b**st**rd!
Right next fit a new one and monitor the bias voltage,those old Mullard ecl’s could be leaking internally.

geoff 1965 wrote: ↑

Thu 12/16/21 4:22 pm

D6D8E25D-70A9-4B8B-BD68-C7376E2613F0.jpeg
Cathode capacitor open! that’s a new sprague atom as well the little b**st**rd!
Right next fit a new one and monitor the bias voltage,those old Mullard ecl’s could be leaking internally.

I've got a NOS ecl86 if you're interested

update: took some time out on the amp to work on my 6V6 project and wait for a batch of toshiba 2968 mosfets I ordered to come in. Decided to remove the tag strips and build a dedicated little turret board for the amp,this started as an experiment really but has turned out to be a cool little amp and definitely a keeper! Left R9 unbypassed for now but room for a capacitor if I want the extra gain.
Filtering is now 50+50+16uf for the power & phase section and 50+30uf for the vvr & preamp,added a sag resistor to the diode rectifier side which will also drop the B+ a bit which was just over 300V on the plates.

New turret board and Toshiba SK2968 mosfet installed,looks a lot better the tag strips are good for experimenting/swapping in and out components but they look bodgey!
Back to fine tuning and finishing off this.

All safety checks done,one tweak to the B+ to preamp replaced the 8K2 dropper with a 6K2 and test voltages are now
V1 pins1&6: 133
V2 pin6: 273. pin1:153
Phase plates: 193&192
Pentodes plate pin6: 288 Screen pin3: 265 Cathode pin7: 9.03
B+ at first cap 306
The cathode voltage on the pentodes is dead stable so that’s a good sign also no DC on the pots.
The 100ohm sag resistor has brought the diode rectified voltages down to virtually the same as tube rectified.
Time to take the dummy load out and plug a G12H 30 in! Just want to experiment with R9 now.
Also the VVR is working perfectly now,very smooth with no crazy fluctuations!

Spoke a bit too soon regarding the VVR! all the test were good under a dummy load,nice smooth drop and increase in voltage when sweeping “but” with the speaker in i’m getting a lot of noise.it’s fine on full or minimum vvr but the noise starts when increasing or decreasing.
The amp is dead quite with no hum and all the lead dress is good,the passive series effects loop is also very good with no noise or oscillations.
Back to the drawing board!
Afterthought I don’t have a spare set of ecl’s at the moment to swap in so not sure if it’s them?