Skip to content

Add pre-transformer gain stage (transformer saturation too subtle, per CTX comparison) #8

Description

@NeoRecasata

Addresses your existing design log note from rev 0.2 review:

"The transformer saturation is very subtle on a mono synth"

Source comparison

I compared the transformer stage against DIYRe CTX Transformer Colour mkII, which is one of the cited design inspirations.

CTX:

  • IN → 100Ω → 0.1µF AC couple → DOA (discrete op-amp) with gain ~3× (R5/R6 = 10K/4.7K) → 249Ω → Cinemag CM-10100 (1:1) → 249Ω load → OUT

Multi-Sat:

  • IN → U1A unity buffer → SW1 → 680Ω → TL016-R (600:600Ω) → HPF → U1B buffer → OUT

The structural difference: CTX adds 3× of pre-transformer gain to drive the magnetic core harder. This Multi-Sat uses only a unity buffer, so the transformer primary sees only the Eurorack-level signal — which is not enough voltage swing to fully saturate the TL016's larger core.

Why this matters

Audio transformers saturate when the magnetic flux exceeds the core's capability. Flux is proportional to volt-seconds at the primary. With 10 Vpp Eurorack levels into 680Ω, the current and resulting flux are modest. CTX's ~3× gain stage puts ~30 Vpp at the primary — well into saturation territory — producing the warm distortion the CTX is known for.

Fix options

A. Add a non-inverting gain stage between U1A and SW1. One spare op-amp section (one TL072 has 2 sections; could reuse an unused half) + two resistors. Specific values to get gain ≈ 3×:

  • R_feedback = 20K to (−) input
  • R_ground = 10K (−) to GND
  • Gain = 1 + 20/10 = 3× ✓

B. Lower R1 from 680Ω to ~200Ω. Pushes more current through the transformer primary without adding components. Mild effect — maybe 1–2 dB more saturation. Quick try first.

C. Switch transformers. TL016 (600:600Ω) → Cinemag CM-10100 or similar (~150Ω). Bigger structural change, higher cost.

I'd recommend B first (one resistor change — try and listen) and A if B isn't enough.

Action

  • Try B in a breadboard test first — change R1 to 220Ω, listen for difference
  • If still too subtle: implement A (add gain stage before transformer)
  • Update calibration procedure: with the new gain, the wet-gain trim (RV2) may need recalibration to keep wet/dry amplitudes balanced

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions