<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>EIGENROOM Field Journal</title>
    <link>https://eigenroom.com/journal</link>
    <atom:link href="https://eigenroom.com/journal/rss.xml" rel="self" type="application/rss+xml"/>
    <description>Practical control engineering notes on PID, frequency design, state space and estimation, with links to reproducible browser experiments.</description>
    <language>en</language>
    <lastBuildDate>Sun, 15 Nov 2026 00:00:00 GMT</lastBuildDate>
    <item><title>Why a lead adds less phase margin than its φmax</title><link>https://eigenroom.com/journal/lead-adds-less-than-phi-max</link><guid isPermaLink="true">https://eigenroom.com/journal/lead-adds-less-than-phi-max</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>FREQUENCY-DOMAIN PRACTICE</category><description>A lead network peaks at one frequency and lifts the loop gain, so the crossover moves to where the plant has less phase. Here is what that costs, in degrees.</description></item>
    <item><title>Mason's formula or block reduction: which is faster?</title><link>https://eigenroom.com/journal/mason-or-block-reduction</link><guid isPermaLink="true">https://eigenroom.com/journal/mason-or-block-reduction</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>MODELING PRACTICE</category><description>Both give the same transfer function. Reduction wins when loops nest or sit in a row; Mason's formula wins when loops overlap or several paths reach the output.</description></item>
    <item><title>Choosing Q and R for LQR without guesswork</title><link>https://eigenroom.com/journal/choosing-lqr-weights</link><guid isPermaLink="true">https://eigenroom.com/journal/choosing-lqr-weights</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>DESIGN DECISIONS</category><description>Q = I and R = 1 mean different things in different units. Bryson's rule turns each weight into an allowed size; a two-state axis shows what rescaling does.</description></item>
    <item><title>Reading a Nyquist plot when the open loop is unstable</title><link>https://eigenroom.com/journal/nyquist-with-an-unstable-open-loop</link><guid isPermaLink="true">https://eigenroom.com/journal/nyquist-with-an-unstable-open-loop</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>STABILITY CHECKS</category><description>With one right-half-plane pole the curve must circle −1 once counter-clockwise, and too little gain is as unstable as too much. Count first, then read margins.</description></item>
    <item><title>Dead time: the limit no gain can buy back</title><link>https://eigenroom.com/journal/dead-time-limits-bandwidth</link><guid isPermaLink="true">https://eigenroom.com/journal/dead-time-limits-bandwidth</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>FREQUENCY-DOMAIN LIMITS</category><description>Delay leaves the gain alone and takes phase in proportion to frequency. On the lab's motor 100 ms costs 19.5° at crossover; only a slower loop wins it back.</description></item>
    <item><title>Discretizing a controller: ZOH, Tustin and the sample period</title><link>https://eigenroom.com/journal/discretizing-a-controller</link><guid isPermaLink="true">https://eigenroom.com/journal/discretizing-a-controller</guid><pubDate>Sun, 15 Nov 2026 00:00:00 GMT</pubDate><category>DIGITAL IMPLEMENTATION</category><description>A sampled loop pays about half a period of delay, and discretization adds its own error. On the lab's motor an 80 ms period costs 41° of phase margin, 10 ms 5°.</description></item>
    <item><title>A smaller z-pole is not always a faster machine</title><link>https://eigenroom.com/journal/a-smaller-z-pole-is-not-always-faster</link><guid isPermaLink="true">https://eigenroom.com/journal/a-smaller-z-pole-is-not-always-faster</guid><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate><category>Measurement and design</category><description>Two logs use different sampling periods. Comparing pole magnitudes without their clocks can reverse your engineering conclusion.</description></item>
    <item><title>Your sensor log can hide an oscillation</title><link>https://eigenroom.com/journal/the-sensor-log-can-hide-an-oscillation</link><guid isPermaLink="true">https://eigenroom.com/journal/the-sensor-log-can-hide-an-oscillation</guid><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate><category>Measurement and design</category><description>A smooth trace can be a sampling artifact. Before retuning a controller, check how the trace was measured.</description></item>
    <item><title>Feedforward does not replace feedback</title><link>https://eigenroom.com/journal/feedforward-does-not-replace-feedback</link><guid isPermaLink="true">https://eigenroom.com/journal/feedforward-does-not-replace-feedback</guid><pubDate>Mon, 05 Oct 2026 00:00:00 GMT</pubDate><category>Measurement and design</category><description>A model can supply an expected command before error develops. The remaining question is what happens when that expectation is wrong.</description></item>
    <item><title>Which error is your feedback actually reducing?</title><link>https://eigenroom.com/journal/which-error-does-feedback-reduce</link><guid isPermaLink="true">https://eigenroom.com/journal/which-error-does-feedback-reduce</guid><pubDate>Sun, 04 Oct 2026 00:00:00 GMT</pubDate><category>Signals in practice</category><description>A sensor gain can make a small summing-junction error coexist with a large physical tracking error.</description></item>
    <item><title>A moving target needs a different tracking test</title><link>https://eigenroom.com/journal/a-moving-target-needs-a-different-test</link><guid isPermaLink="true">https://eigenroom.com/journal/a-moving-target-needs-a-different-test</guid><pubDate>Sun, 04 Oct 2026 00:00:00 GMT</pubDate><category>Design practice</category><description>Zero step error does not guarantee zero ramp error. Use the input your application actually needs.</description></item>
    <item><title>Positive coefficients are not a stability certificate</title><link>https://eigenroom.com/journal/positive-coefficients-are-not-enough</link><guid isPermaLink="true">https://eigenroom.com/journal/positive-coefficients-are-not-enough</guid><pubDate>Sun, 04 Oct 2026 00:00:00 GMT</pubDate><category>Stability checks</category><description>A cubic with positive coefficients can still have two growing modes.</description></item>
    <item><title>Why the fastest observer is not always the best</title><link>https://eigenroom.com/journal/fast-observers-have-a-cost</link><guid isPermaLink="true">https://eigenroom.com/journal/fast-observers-have-a-cost</guid><pubDate>Sat, 03 Oct 2026 00:00:00 GMT</pubDate><category>Design practice</category><description>If the observer trusts each noisy position reading more strongly, what happens to the control effort?</description></item>
    <item><title>One tank, many linear models</title><link>https://eigenroom.com/journal/one-tank-many-linear-models</link><guid isPermaLink="true">https://eigenroom.com/journal/one-tank-many-linear-models</guid><pubDate>Sat, 03 Oct 2026 00:00:00 GMT</pubDate><category>Design practice</category><description>Does the water level return on the same time scale at low and high operating heights?</description></item>
    <item><title>Which RLC resonance did you measure?</title><link>https://eigenroom.com/journal/which-rlc-resonance</link><guid isPermaLink="true">https://eigenroom.com/journal/which-rlc-resonance</guid><pubDate>Sat, 03 Oct 2026 00:00:00 GMT</pubDate><category>Design practice</category><description>Must the largest capacitor-voltage response occur exactly at ωn = 1/√(LC)?</description></item>
    <item><title>A heater cannot cool: check authority before tuning</title><link>https://eigenroom.com/journal/a-heater-cannot-cool</link><guid isPermaLink="true">https://eigenroom.com/journal/a-heater-cannot-cool</guid><pubDate>Sat, 03 Oct 2026 00:00:00 GMT</pubDate><category>Actuators &amp; design</category><description>Use a thermal example to separate slow dynamics, impossible targets and integral windup. A better gain cannot create an actuator you do not have.</description></item>
    <item><title>Why more damping can make a system slower</title><link>https://eigenroom.com/journal/damping-is-not-always-faster</link><guid isPermaLink="true">https://eigenroom.com/journal/damping-is-not-always-faster</guid><pubDate>Sat, 03 Oct 2026 00:00:00 GMT</pubDate><category>PHYSICAL INTUITION</category><description>Use a mass–spring–damper to separate overshoot, settling and energy loss. Compare underdamped, critical and overdamped responses before treating damping as a universal speed knob.</description></item>
    <item><title>Do you need feedback if a voltage already gives the right speed?</title><link>https://eigenroom.com/journal/open-loop-versus-feedback</link><guid isPermaLink="true">https://eigenroom.com/journal/open-loop-versus-feedback</guid><pubDate>Fri, 02 Oct 2026 00:00:00 GMT</pubDate><category>FEEDBACK FOUNDATIONS</category><description>Compare a held voltage with speed feedback under the same load. Learn when open-loop control is useful, what feedback changes, and why an open-loop Bode plot is a different idea.</description></item>
    <item><title>Why a spring and an RLC circuit share an equation</title><link>https://eigenroom.com/journal/mass-spring-damper-and-rlc</link><guid isPermaLink="true">https://eigenroom.com/journal/mass-spring-damper-and-rlc</guid><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>MODELING ACROSS DOMAINS</category><description>Match mass, damping and stiffness to inductance, resistance and inverse capacitance—without confusing an analogy with an actual motor circuit.</description></item>
    <item><title>When does a linearized cart-pole model stop helping?</title><link>https://eigenroom.com/journal/when-linearization-breaks</link><guid isPermaLink="true">https://eigenroom.com/journal/when-linearization-breaks</guid><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>MODEL LIMITS</category><description>A controller designed near upright may fail after a larger tilt, but a failed run alone does not measure the approximation error.</description></item>
    <item><title>A calm-looking trace is not proof of stability</title><link>https://eigenroom.com/journal/short-run-is-not-stability</link><guid isPermaLink="true">https://eigenroom.com/journal/short-run-is-not-stability</guid><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>EVIDENCE AND STABILITY</category><description>A short simulation checks one input and one initial condition; BIBO and internal stability make broader claims.</description></item>
    <item><title>Write a control specification before tuning gains</title><link>https://eigenroom.com/journal/write-a-control-specification</link><guid isPermaLink="true">https://eigenroom.com/journal/write-a-control-specification</guid><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>DESIGN WORKFLOW</category><description>Turn 'fast and stable' into observable requirements for tracking, load rejection, actuator effort and limits.</description></item>
    <item><title>Why delay and right-half-plane zeros resist faster control</title><link>https://eigenroom.com/journal/delay-and-nonminimum-phase-limits</link><guid isPermaLink="true">https://eigenroom.com/journal/delay-and-nonminimum-phase-limits</guid><pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate><category>FREQUENCY-DOMAIN LIMITS</category><description>Two different mechanisms—late information and an initial wrong-way response—can limit useful bandwidth and make aggressive tuning backfire.</description></item>
    <item><title>Can you tune PID without a plant model?</title><link>https://eigenroom.com/journal/pid-without-a-model</link><guid isPermaLink="true">https://eigenroom.com/journal/pid-without-a-model</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>DESIGN DECISIONS</category><description>Yes—but model-free does not mean evidence-free. A safe experiment still needs measured signals, limits and repeatable conditions.</description></item>
    <item><title>PID or state-space? First fix the question.</title><link>https://eigenroom.com/journal/state-space-or-pid</link><guid isPermaLink="true">https://eigenroom.com/journal/state-space-or-pid</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>DESIGN DECISIONS</category><description>PID is a controller family. State-space is a model representation. LQR is one design method built on that representation.</description></item>
    <item><title>Lead or lag? Read the Bode plot, then check the motor.</title><link>https://eigenroom.com/journal/lead-lag-from-bode-to-motor</link><guid isPermaLink="true">https://eigenroom.com/journal/lead-lag-from-bode-to-motor</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>FREQUENCY-DOMAIN PRACTICE</category><description>Phase near crossover and low-frequency accuracy are different goals. A good frequency plot can still hide actuator saturation.</description></item>
    <item><title>A smooth Kalman estimate is not yet a good controller.</title><link>https://eigenroom.com/journal/kalman-filter-to-lqg</link><guid isPermaLink="true">https://eigenroom.com/journal/kalman-filter-to-lqg</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>ESTIMATION IN PRACTICE</category><description>Separate state estimation from feedback, test your noise assumptions, then compare LQG with ideal full-state LQR.</description></item>
    <item><title>Why does a motor overshoot with a healthy phase margin?</title><link>https://eigenroom.com/journal/phase-margin-but-bad-response</link><guid isPermaLink="true">https://eigenroom.com/journal/phase-margin-but-bad-response</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>DESIGN DIAGNOSIS</category><description>A linear margin describes one loop model near an operating condition; a limited actuator can tell a different time-domain story.</description></item>
    <item><title>Lead, lag or PI: which problem are you solving?</title><link>https://eigenroom.com/journal/lead-lag-or-pi</link><guid isPermaLink="true">https://eigenroom.com/journal/lead-lag-or-pi</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>CONTROLLER CHOICES</category><description>Compare phase, persistent error and actuator effort under the same motor target and load before choosing a compensator.</description></item>
    <item><title>One good step fit does not validate a motor model.</title><link>https://eigenroom.com/journal/model-fit-is-not-validation</link><guid isPermaLink="true">https://eigenroom.com/journal/model-fit-is-not-validation</guid><pubDate>Sat, 26 Sep 2026 00:00:00 GMT</pubDate><category>MODELING PRACTICE</category><description>Parameter fitting and held-out prediction answer different questions; use a second target or load trial before trusting a design model.</description></item>
  </channel>
</rss>
