Update: See this new post on what I turned this build in to. This build had too much gain and was too unstable.
Goal: The ultimate living room amp. Guitar + cable + amp. No pedals. Enough distortion to cover 90s rock / grunge.
Background: After building my AmpMaker N5X, a lovely amp in itself, I still had the need for more gain. I was ending up using a boost or overdrive/distortion pedal in front to get a heavier sound. So I set about adding an extra 2 gain stages, creating a “N5X2“.
Outcome: Below are some pictures of the build and a sound clip. A bit more tinkering may be needed to get the best out of it. But it sounds great to me.
Recorded using a US Std Strat (with Seymour Duncan Everything Axe pickups: mini humbuckers neck and bridge, and a duckbucker in the middle), straight into the amp. Speaker is a 12″ Tayden True Brit. Mic’d using a Samson USB condenser mic, straight in to GarageBand. No EQ / effects / compression etc.
Valves used were 2 Sovtek ECC83s (borrowed from my old Hot Rod Deluxe) and a JJ EL84.
G1 / G2 on full, MV low, Power at minimum. Treble on full, Mid on 3, Bass on 7. N5X boost switch set to bright – 1uF bypass caps.
00:00 – 00:30 Guitar vol on ~2 or 3.
00:30 – 00:49 vol on ~6/7
00:49 – 01:30 vol on full
Middle – Duckbucker
01:30 – 01:43 vol on full
01:43 – 02:11 vol on 6ish
02:10 – 02:50 vol on 3ish
02:50 – 03:07 vol on full
03:07 – 03:22 vol on full
G1 / G2 on 8, MV on full, Power at minimum
03:25 – 03:35 vol on 2 or 3
03:35 – 03:44 vol on 6
03:44 – 04:03 vol on full
The clip below is recorded with everything up full, apart from Treble and Mids which are zero.
This clip is with the tone stack bypassed. I turn up the MV to full then I switch between high boost, boost off and low boost – switching the bypass caps on the 3rd and 4th stages.
Update: This is seemingly solved by altering my circuit – decoupling the power supply nodes for the 2 preamp valves… Updated schematic with my measured voltages is below.
Below are some images generated using a software spectrum analyser / scope.
Bias Calculations – JJ valves & 1 Mullard
Below are some graphs I generated to try and establish the operating points for each stage and then an estimation of the harmonic distortion (H).
For each preamp stage my measured voltages are off by ~5-8%. However, looking at this and on the advice given on the AX84 forum, such deviation is to be expected.
For the output stage I scaled the curves to match my readings.
Stages 1 and 2
Updated Bias Calculations – All Mullard
I got myself a pair of NOS Mullards. Finally, my readings match closely to the charts! Other than the EL84 readings… But it is an old valve, also Mullard, but not NOS.
Stages 1 and 2
For guidance and inspiration I used the resources at AX84.com. I initially focussed on the High Octane design (3 stages + cathode follower), then expanded to look at the L2L schematic (4 stages). The advice given was to simply bolt the extra stages to the front of the stock N5X. This pushed me in the direction of the L2L.
The L2L’s main selling point to me was the apparent ability to use the guitar volume control to go from clean to mean. This stems from the anode feedback circuit around the 2nd stage. The design is also close to Merlin’s UHG circuit. So I should be able to switch between the two should I wish. I also wanted to be able to return to the stock N5X circuit for lower-gain sounds.
I spent a great deal of time working out the simplest and best way to extend the N5X kit to these circuits, with lots of help from the AX84 forum. Below is what I came up with and a brief description of the build.
Changes to N5X
The main changes are:
- Elevated heaters to avoid noise
- Restructured grounding layout to follow Merlin’s guidance to hopefully avoid noise and parasitic oscillations etc
- 2 extra gain stages
These required the original power amp board to be modified and a new preamp board to be created. New components required are listed below, and 99.9% of the original kit is used. I’ll be making changes to the schematic as I tinker to get the best out of the amp. I’ll keep uploading the latest version as I go.
This is what I came up with. It takes Barry’s existing N5X schematic and meets the L2L or UHG in the middle. The new areas are highlighted in pink. The use of screened wires is highlighted in yellow. The design has the capability of returning to the existing N5X circuit (switch S4).
Update 10/11/15: I’ve changed a few component values through testing – dropping the power to the 1st valve helped with oscillations, adding in an extra signal attenuation resistor between stage 2 and 3 helps prevent too much clipping in the 3rd stage, and increasing the lower resistor of the voltage divider seems to increase the negative feedback. I also lowered the first two coupling capacitors to 1nF.
This board fits on the existing mounting holes. All that is required is to drill an extra 3 holes for the 3 extra turrets. One is used for the improved grounding scheme. The other two are for the elevated heaters (improving my earlier modification). The unused wired from the transformer that were connected to the turrets on the right need to be safely terminated / insulated. Perhaps by folding them individually back on themselves and covering with shrink wrap or insulating tape.
This board also fits on the existing mounting holes. All that is required is the turret board material, turrets, drill bits of the correct size and a turret staking tool. All this is available from AmpMaker. Print out the layout at 100%, afix to the board of the correct size, drill the holes of the correct size where indicated, then insert / splay the turrets.
I’ve altered the design to keep the filter capacitors close to the preamp stage they supply. This comes from the improved grounding scheme developed.