Predictika · Power supply selection

An AI sales agent's power supply selection, checked before it goes out

An AI sales agent recommends a 10 W AC-DC power supply from the ECE10 series for a customer's design: the model for the output voltage, the mounting style, the input supply, the continuous and peak load, the ambient temperature and the nominal current for the quote. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the manufacturer's public datasheet and returns a verdict, every reason and the figures it computed.

Scope: Selections from the ECE10 series (ECE10US03 to ECE10US48, encapsulated or open frame): output voltage, input, load with derating, peak load, ambient temperature, altitude and nominal current. See everything this validator covers

This validator models selections from the ECE10 series, not just the scenarios shown below.

  • Models — the seven output voltages 3.3, 5, 9, 12, 15, 24 and 48 V; a model that does not deliver the voltage the design needs is blocked.
  • Input — 85 to 264 VAC at 47 to 63 Hz, or 120 to 370 VDC; the load is derated for AC input between 85 and 90 V.
  • Load and ambient — the continuous load is compared with the nominal current after derating for low AC input and for ambient temperature above 50 C (to 50 percent at 70 C); the peak load must stay within the peak rating and last under 30 seconds at a duty cycle of 10 percent or less; ambient temperature from -25 to +70 C and altitude up to 3,048 m.
  • Mounting and stated current — the open frame version (suffix -P) is for OEM quantities and goes to a person for confirmation; the nominal current on the quote is compared with the datasheet figure.

Any selection from this series that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (the datasheet lists none, so this validator produces no price build-up), the quantity at which open frame becomes available, other XP Power series, and approvals or EMC suitability for a specific end product.

The AI sales agentSelects a power supply from the customer's requirements and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      
Waiting for the call.

Where rules-based CPQ gets it wrong

Click to expand

Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

What the validator caughtTypical rules-based CPQPredictika
A usable load that depends on input voltage and temperatureOnly with workarounds
The current available falls at low AC input and again above 50 C; the two derating curves multiply, so the rule has to be built as a computed value and then compared with the load.
The derated current is computed from the model, input and ambient temperature on every call and compared with the load.
An input range that depends on AC or DCOnly with care
AC and DC have separate windows and only AC has a frequency range; a rule per case is workable but easy to miss.
The input type, voltage and frequency are checked together.
A peak load with two limitsOnly with care
A peak must be under the peak rating, shorter than 30 seconds and infrequent enough; three conditions that have to be remembered together.
The peak rating, duration and duty cycle are all checked and each failure is reported.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the selection is saved or a step is completed, so a rep often fixes one error and meets the next.
All findings come back together, whatever order the choices arrived in.

What this validator catches: findings from its test cases

Click to expand

Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the quote; approval findings need a person to sign off; everything else is computed.

1The model on the quote does not deliver the required output voltage; the required output is 24 V and the model gives 12 V.
2The AC input of 80 V is outside the 85 to 264 VAC range of the ECE10 series.
3The AC input of 270 V is outside the 85 to 264 VAC range of the ECE10 series.
4The DC input of 100 V is outside the 120 to 370 VDC range of the ECE10 series.
5The DC input of 400 V is outside the 120 to 370 VDC range of the ECE10 series.
6The input frequency of 40 Hz is outside the 47 to 63 Hz range of the ECE10 series.
7The input frequency of 400 Hz is outside the 47 to 63 Hz range of the ECE10 series.
8The ambient temperature of -30 C is outside the -25 to +70 C operating range.
9The ambient temperature of 75 C is outside the -25 to +70 C operating range.
10The operating altitude of 4,000 m is above the 3,048 m the ECE10 series is specified for.
11The continuous load of 800 mA is above the 747 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).
12The continuous load of 700 mA is above the 622.5 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).
13The continuous load of 700 mA is above the 415 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).
14The continuous load of 500 mA is above the 373.5 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).
15The peak load of 1,200 mA is above the 1,000 mA peak rating of this model.
16The peak load must last less than 30 seconds with a duty cycle of 10 percent or less; the quote has 30 seconds at 5 percent.
17The peak load must last less than 30 seconds with a duty cycle of 10 percent or less; the quote has 10 seconds at 15 percent.
18The quote states a nominal output current of 1,000 mA, but the datasheet lists 830 mA for this model.
19The open frame version (suffix -P) is available for OEM quantities only; the datasheet gives no quantity, so a person must confirm the order qualifies.
20The model on the quote does not deliver the required output voltage; the required output is 5 V and the model gives 12 V.
21The model on the quote does not deliver the required output voltage; the required output is 12 V and the model gives 24 V.
22The AC input of 300 V is outside the 85 to 264 VAC range of the ECE10 series.
23The input frequency of 30 Hz is outside the 47 to 63 Hz range of the ECE10 series.
24The ambient temperature of 80 C is outside the -25 to +70 C operating range.
25The operating altitude of 5,000 m is above the 3,048 m the ECE10 series is specified for.
26The continuous load of 900 mA is above the 102.5 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).
27The peak load of 600 mA is above the 490 mA peak rating of this model.
28The peak load must last less than 30 seconds with a duty cycle of 10 percent or less; the quote has 40 seconds at 20 percent.
29The quote states a nominal output current of 500 mA, but the datasheet lists 410 mA for this model.

Scenarios and results

Click to expand

The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

1. 12 V model for a 24 V design: An ECE10US12 for a 24 V design on a 270 VAC supply with a 1,200 mA peak

The agent's request. Please quote the ECE10US12 power supply, the encapsulated version, for a design that needs 24 V at 350 mA continuous and a peak load of 1200 mA for 20 seconds at a 5 percent duty cycle, powered from 270 VAC at 50 Hz in an ambient temperature of 25 C. The quote states a nominal output current of 830 mA. It is ready to send.

Verdict: Blocked.

  • Model voltage differs from the required output. The model on the quote does not deliver the required output voltage; the required output is 24 V and the model gives 12 V.
  • AC input outside 85 to 264 VAC. The AC input of 270 V is outside the 85 to 264 VAC range of the ECE10 series.
  • Peak current above the peak rating. The peak load of 1,200 mA is above the 1,000 mA peak rating of this model.

2. 12 V model for a 24 V design: The agent resubmits after fixing the errors

The agent's request. Please quote the ECE10US24 power supply, the encapsulated version, for a design that needs 24 V at 350 mA continuous and a peak load of 450 mA for 20 seconds at a 5 percent duty cycle, powered from 230 VAC at 50 Hz in an ambient temperature of 25 C. The quote states a nominal output current of 410 mA. It is ready to send.

Verdict: Allowed.

  • Selection summary. Nominal output 410 mA at the model voltage, 410 mA available after derating, peak rating 490 mA, efficiency 83 percent at 115 VAC.

3. Hot enclosure at low line: An ECE10US05 carrying 1,700 mA at 85 VAC in a 65 C enclosure

The agent's request. Please quote the ECE10US05 power supply, the encapsulated version, for a design that needs 5 V at 1,700 mA continuous, powered from 85 VAC at 50 Hz in an ambient temperature of 65 C. The quote states a nominal output current of 2,000 mA. It is ready to send.

Verdict: Blocked.

  • Load above the derated current. The continuous load of 1,700 mA is above the 1,125 mA this model can deliver at the input voltage and ambient temperature on the quote (nominal current derated for low AC input and for temperature above 50 C).

4. Hot enclosure at low line: The agent resubmits after fixing the errors

The agent's request. Please quote the ECE10US05 power supply, the encapsulated version, for a design that needs 5 V at 1,700 mA continuous, powered from 85 VAC at 50 Hz in an ambient temperature of 45 C. The quote states a nominal output current of 2,000 mA. It is ready to send.

Verdict: Allowed.

  • Selection summary. Nominal output 2,000 mA at the model voltage, 1,800 mA available after derating, peak rating 2,400 mA, efficiency 80 percent at 115 VAC.

5. 48 V model on a low DC bus: An ECE10US48 on a 100 VDC bus with a stated nominal current of 250 mA

The agent's request. Please quote the ECE10US48 power supply, the encapsulated version, for a design that needs 48 V at 180 mA continuous, powered from 100 VDC in an ambient temperature of 25 C. The quote states a nominal output current of 250 mA. It is ready to send.

Verdict: Blocked.

  • DC input outside 120 to 370 VDC. The DC input of 100 V is outside the 120 to 370 VDC range of the ECE10 series.
  • Stated nominal current differs. The quote states a nominal output current of 250 mA, but the datasheet lists 210 mA for this model.

6. 48 V model on a low DC bus: The agent resubmits after fixing the errors

The agent's request. Please quote the ECE10US48 power supply, the encapsulated version, for a design that needs 48 V at 180 mA continuous, powered from 300 VDC in an ambient temperature of 25 C. The quote states a nominal output current of 210 mA. It is ready to send.

Verdict: Allowed.

  • Selection summary. Nominal output 210 mA at the model voltage, 210 mA available after derating, peak rating 250 mA, efficiency 83 percent at 115 VAC.

7. Long peak load: An ECE10US15 with a 700 mA peak lasting 40 seconds

The agent's request. Please quote the ECE10US15 power supply, the encapsulated version, for a design that needs 15 V at 500 mA continuous and a peak load of 700 mA for 40 seconds at a 5 percent duty cycle, powered from 230 VAC at 50 Hz in an ambient temperature of 25 C. The quote states a nominal output current of 660 mA. It is ready to send.

Verdict: Blocked.

  • Peak load too long or too frequent. The peak load must last less than 30 seconds with a duty cycle of 10 percent or less; the quote has 40 seconds at 5 percent.

Narration transcript

Click to expand

The text of the narrated walkthrough, step by step.

  1. Welcome to a demo of Predictika's Validator. An AI sales agent recommends a 10 W AC-DC power supply from the ECE10 series for a customer's design: the model for the output voltage, the mounting style, the input supply, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the manufacturer's public datasheet and returns a verdict, every reason and the figures it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please quote the ECE10US12 power supply, the encapsulated version, for a design that needs 24 V at 350 mA continuous and a peak load of 1200 mA for 20 seconds at a 5 percent duty cycle. The quote states a nominal output current of 830 mA. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with three findings, all in one answer. The model on the quote does not deliver the required output voltage; the required output is 24 V and the model gives 12 V. The AC input of 270 V is outside the 85 to 264 VAC range of the ECE10 series. The peak load of 1,200 mA is above the 1,000 mA peak rating of this model.
  6. The validator returned three blocking findings in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the quote are checked against each other.
  7. What this shows: for a usable load that depends on input voltage and temperature, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the model to ECE10US24, the input voltage to 230 V, the peak load to 450 mA and the stated nominal current to 410 mA.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. All three earlier findings are cleared, and the validator's figure matches the quote.
  13. Three problems caught in one call, fixed in one pass, and a clean quote. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.