Troubleshooting

Most first-session surprises on this tool are one of three things: a setup error that has a direct and named fix, a misread that looks like a malfunction and resolves when you read the pane in the right order, or a ge...

Written By Axiom Admin

Last updated 22 days ago

Troubleshooting

Most first-session surprises on this tool are one of three things: a setup error that has a direct and named fix, a misread that looks like a malfunction and resolves when you read the pane in the right order, or a genuine limit that cannot be "fixed" because the tool is behaving correctly and the configuration is the constraint.

This page is organized around those three categories, in that order. The goal is to teach you to triage rather than to memorize fixes β€” once you can tell which category a symptom belongs to, the specific fix (or specific acceptance) becomes a short step rather than a long hunt.

A working rule for triage. When something looks wrong, ask in this order: is TradingView showing an error, is the pane empty, or is the pane drawing but reading strangely? The first points at a setup error nine times out of ten. The second points at an enablement or warm-up problem. The third is almost always a misread β€” the tool is running, the chart is real, and the gap is between what you are seeing and what you expected to see. Work the categories in that order and you will spend less time hunting for a fix that is not needed because the tool is not the problem.

Reading the categories

  • Setup errors are symptoms where the tool or the chart has been configured in a way that prevents it from running or from running meaningfully. The fix is a configuration change.

  • Misreads are symptoms where the tool is running correctly but the reader is interpreting the output against the wrong mental model. The fix is a reading change, not a configuration change. These are the most common first-week surprises.

  • Genuine limits are symptoms where the tool is running correctly, the reader is reading correctly, and the configuration is nonetheless not producing the effect the reader wanted. The fix is either to accept the behavior, to change the configuration to something that fits, or to choose a version whose surface actually carries that requirement.

Setup errors

Runtime error naming a slot

Symptom. The pane is blank or broken and TradingView is showing a runtime error that reads like RSI 02 timeframe cannot be lower than the chart timeframe.

Likely cause. The named slot's timeframe is below the chart's timeframe. The tool raises this error deliberately β€” a slot timeframe below the chart timeframe has no coherent interpretation here.

Check. Open the indicator settings. Find the named slot. Check its TimeFrame: value against the chart's current timeframe.

Fix. Set the slot's timeframe to a value at or above the chart's timeframe, or clear the slot's timeframe field to inherit the chart's own. Alternatively, switch the chart to a timeframe below the slot's configured timeframe.

Pane is uniform across all three slots

Symptom. The three slot lines sit on top of each other, the blend tracks the slots exactly, and the pane looks like one RSI read.

Likely cause. Same-timeframe collapse or duplicated slot settings. The script does not force lower slot timeframes upward; it errors when a slot timeframe is below the chart timeframe. Uniform panes usually mean the slots have been set to the same timeframe, left blank to inherit the chart timeframe, and configured with similar settings.

Check. Compare each slot's timeframe, source, RSI length, MA family, RSI smoothing, and signal smoothing. If those match, the slots are answering the same question and the pane should look uniform.

Fix. Rebuild the stack deliberately: keep every slot timeframe greater than or equal to the chart timeframe, and give the slots distinct jobs if you want distinct reads. If TradingView shows the lower-timeframe runtime error, raise the offending slot timeframe or lower the chart timeframe.

Nothing plots, no error

Symptom. The pane is empty. No runtime error. The reference lines draw.

Likely cause 1. All three slots are disabled. Likely cause 2. All three slots are enabled but still in warm-up on a chart with very little history. Likely cause 3. The blend display is off (Plot Blended RSI/Signal = false) and every slot has Hide Plot on.

Check. Open settings. Confirm at least one slot is enabled and not hidden. Confirm the chart has enough history for the slowest slot's RSI length plus both smoothing passes to fill.

Fix. Enable a slot. Un-hide at least one plotted slot or turn the blend display on. Scroll the chart back to a period with more history if warm-up is the cause.

Pane crowded and unreadable

Symptom. Too many lines overlapping, full-tone and faded colors blending into a visual noise that makes the reading order impossible to apply.

Likely cause. Three slots at default line width plus the blend at default blend width on a short chart pane. Master smoothing can change the blend's shape, but it does not add extra plotted lines.

Check. The TradingView pane height, the line widths, hidden-slot settings, and whether the blend itself needs to be hidden for the read you are trying to make.

Fix. Reduce slot line widths, increase the pane height, or use Hide Plot on slots you want to keep running but not see (remembering they still feed the blend and alerts). See the Hide Plot section of Settings for the tradeoff.

Misreads

Slot color is "wrong" relative to the value

Symptom. A slot line sits at 60 but the color is faded. A slot line sits at 45 but the color is full-tone. The value and color seem to disagree.

Likely cause. This is not a disagreement. The color comes from RSI vs signal, not RSI vs 50. The slot's smoothed RSI at 60 faded means the RSI just fell below the signal line even though the value is still above the midline. The slot at 45 full-tone means the RSI has climbed above its signal even though the value is below the midline.

Check. Read the Visuals and Logic page β€” specifically the section on reading color first and value second.

Fix. Not a fix; a reading correction. The color and value encode different things, and the pane is designed around keeping both visible at once. This is one of the most common first-week misreads, and it resolves as soon as the reading order installs.

"All three slots agree" not producing the breadth read it looked like

Symptom. The All RSI Slots Bullish or All RSI Slots Bearish alignment alert fires frequently, sometimes for long runs of chart bars at a time, and the "agreement" does not feel like it carries much information.

Likely cause. Most commonly, the three slots are running identical settings and differ only by timeframe. The alignment is one measurement sampled at three cadences rather than three independent observations. A trending instrument will keep such an alignment holding for extended runs, and the alignment alert re-fires on every confirmed chart bar.

Check. Open settings and compare the three slots. If source, RSI length, MA family, and both smoothing lengths are identical across slots, you have an identical-settings stack.

Fix. Differentiate the slots. Give each slot a job (see the differentiated three-slot workflow in Workflows). Differentiation reduces the autocorrelation; it does not eliminate it. Limitations and Trust Boundaries carries the full treatment of why alignment is a cadence-coverage read rather than a breadth read.

A per-slot alert fires on every chart bar for a long run

Symptom. You wired RSI 01 Is Bullish as a once-per-bar alert. It is firing on every single chart bar for thirty bars straight.

Likely cause. The alerts are state-based, not edge-triggered. If the slot holds its bullish state across many chart bars, the bullish alert re-fires on every confirmed chart bar where the state holds.

Check. The slot's color on the pane β€” is the slot sitting in full-tone color for thirty bars? If yes, the alert is firing on the held state, which is correct behavior.

Fix. Not a fix; a design detail. The tool does not ship native transition alerts. Blend-level plotted lines can support some TradingView custom alert work, but per-slot RSI-vs-signal transitions are not exposed as two plotted lines because the per-slot signal is hidden.

Blend fired an alert but the blend is hidden

Symptom. You turned Plot Blended RSI/Signal off. The blend visuals are gone. A blend-based alert (Blended RSI Is Bullish or Blended RSI Is Bearish) still fired.

Likely cause. The display toggle hides the blend visuals but does not stop the blend from computing. Blend-based alerts evaluate on the blend state regardless of whether the blend is drawn.

Check. The Plot Blended RSI/Signal input and the wired blend alerts.

Fix. Not a fix; a design detail. If you do not want blend alerts firing, remove the wired alerts. The separation between display and logic is deliberate.

A slot is clearly influencing the blend but the slot line is not on the pane

Symptom. The blend is moving in ways that cannot be explained by the two slot lines you can see. One slot line is missing.

Likely cause. The missing slot has Hide Plot on. The slot is still computing, still contributing to the blend at its weight, and still feeding its own alerts.

Check. Open settings and check Hide Plot for each slot. Also check the slot's Enable state β€” a disabled slot does not plot, does not blend, and does not alert, and the symptom does not arise from a disabled slot.

Fix. If the hidden slot should be fully out of the blend, set its weight to zero (and remember zero weight still counts for alignment alerts) or disable it outright. If the hidden slot is intentional β€” you wanted the blend contribution without the extra line on the pane β€” the behavior is correct and the slot is doing its job.

A slot at weight zero is still producing alerts

Symptom. You set a slot's weight to zero and wired no blend alerts on it, but you are still getting per-slot alerts for that slot, or you are seeing the slot counted in the alignment alerts.

Likely cause. Weight zero excludes the slot from the blend only. The slot still computes, still plots (unless Hide Plot is on), still feeds per-slot alerts, and still counts toward alignment.

Check. Confirm the slot's Enable state, weight, and Hide Plot setting. Confirm which alerts are wired on the slot.

Fix. If you want the slot completely inactive, set Enable off. If you want only the blend excluded, the current configuration is correct and the alerts are expected.

Genuine limits

No alert for a crossover, divergence, or threshold event

Symptom. You are looking for an alert condition on a crossover between RSI and signal, or on a 70-line touch, or on a divergence between price and RSI, and the list of conditions does not include one.

Likely cause. Not a bug. The trim does not ship transition alerts, threshold alerts on 30/70/0/100, or divergence alerts. The available conditions are ten state and state-alignment conditions.

Check. Alerts enumerates the ten conditions and the named absences.

Fix. If you need a threshold alert on a plotted RSI line, TradingView can usually build that from the plotted series. If you need per-slot RSI-vs-signal transition alerts, this trim does not expose the hidden per-slot signal as a plotted line, so that condition is not available through configuration. The tool does not ship these conditions natively, and you will not find them by changing inputs.

I want per-slot repaint control

Symptom. You want On Bar Close? = true on slot 03 and On Bar Close? = false on slot 01.

Likely cause. Not supported on this trim. On Bar Close? is a single global switch. Per-slot repaint control is on CTX.

Check. The PU Settings group in the indicator's input dialog.

Fix. Either choose a single repaint posture for all three slots, or move to CTX if ten-slot context breadth with per-slot control is the configuration you need. STR also carries per-slot control, but it is the structure version: five slots plus extra reads around the blended RSI output.

I want a Stoch-RSI

Symptom. You want the pane to be running a Stochastic RSI with the %K and %D lines in the 0..1 band.

Likely cause. Not what this tool is. Base leaves RSI in its native 0..100 range and smooths it twice; Stoch-RSI rescales RSI into a 0..1 band before smoothing. Different construction, different pane.

Check. For the Geeks carries the structural comparison.

Fix. Not a configuration fix. A Stoch-RSI is a different instrument.

Master smoothing "added noise" instead of calming the blend

Symptom. You enabled master smoothing hoping to calm the blend; the blend is now calmer but you are also noticing blend-based alerts firing later than you expected, and the blend is flattening through moves you can see on the slot lines.

Likely cause. Working as designed. Master smoothing trades responsiveness for calm. Long master lengths on top of long per-slot signal smoothing can flatten the blended pair right through a regime change that is still visible on the slots.

Check. The master length and the per-slot signal lengths. If both are long, the effective blend lag is the combination.

Fix. Either shorten the master length, shorten the per-slot signal lengths, or disable master smoothing. A noisy blend is evidence that the slots are disagreeing β€” smoothing away the evidence does not resolve the disagreement.

The color on one slot flickers rapidly on fast bars

Symptom. A slot's color flips full-tone to faded and back over a handful of chart bars. The pane feels jittery even though the rest of the slots are calm.

Likely cause. On a fast instrument or a news bar, the smoothed RSI and the smoothed signal can cross and re-cross across a short span. That is the slot reporting real, short-lived direction changes; it is not a malfunction.

Check. The slot's smoothing lengths and the instrument's realized volatility during the window. A length-3 RSI smoothing with a length-3 signal on a news-driven 5m chart will flicker more than the same slot during a calm session.

Fix. If the flicker is costing you more than the responsiveness is earning, raise the RSI smoothing or the signal length (or both) on the flickering slot. Accept the lag that trade brings. If the flicker only bothers you on specific sessions, consider running the flickery configuration during calm periods and a calmer configuration during fast ones β€” but pick a posture and commit rather than flipping mid-session.

Where to go next