Know Your Water

EPA Standards Miss Many Chemicals in Drinking Water, Study SaysMore

TDSBot for RO & other purifier manufacturers

OEM IoT Intelligence for Reverse Osmosis Systems

A dual-sensor TDSBot embeds in your RO product line and reports reject ratio continuously from every deployed unit — field performance data your R&D, sales and OEM teams can each act on, plus a recurring SaaS revenue line.

RO system · intelligence built in at manufactureRO system · intelligence built in at manufacture
Membrane performance monitorAlert below 82%
Tap a tab above to see readings change
TDS_inlet420ppm · before membrane
TDS_outlet21ppm · permeate / treated
Inlet → Membrane → Permeate
Reject ratio95.0%
Membrane performing normally. Reject ratio 95.0% — well above 82% alert threshold.
Active MOU · leading RO manufacturerCommercial RO pilot · in deploymentHigh-volume OEM discussion · ongoing
For Product & R&D
Your membrane performs to spec in the lab. What happens after it ships?

Your warranty returns tell you something failed. They don't tell you when, at what water chemistry, or whether it was a product failure or an installation failure. TDSBot gives your R&D team continuous field performance data — reject ratio over the membrane's lifetime, across geographies, at scale.

Real-time
Reject ratio from every deployed unit
400+
Deployment geographies tracked in pilot
For Technical Sales
Your competitors quote the same spec sheet. You can show the real-world data.

When a distributor asks why your membrane costs more, "98% rejection under lab conditions" is a weak answer. A 24-month reject ratio dataset from 400 real installations — continuous, timestamped, across water chemistries — is not. TDSBot turns your installed base into a living proof point.

24 mo
Field performance record per unit
3
Failure modes distinguished by dual-sensor data
For OEM & Partnerships
You ship 50,000 filters a year. How many are performing right now?

After installation, your visibility ends. A warranty return is the first signal that something went wrong — weeks after the fact. TDSBot embeds into your product line and gives you continuous performance visibility across your entire installed base — and a recurring SaaS revenue stream that grows with every unit shipped.

3
Different monitoring plans, OEM discounts available
10K+
SaaS revenue scales with deployments
Why OEMs choose WaterTDS

Four outcomes.
One platform underneath.

Before the sensor architecture, the reason manufacturers embed it.

Differentiate
Launch a connected product line

Compete on data, not just spec sheets. A 24-month field performance dataset from real installations is a claim your competitors quoting lab-only numbers can't match.

Reduce warranty costs
Know the fault before you dispatch service

Dual-sensor bypass valve drift detection distinguishes a membrane fault from a valve fault in seconds — instead of replacing membranes that were never the problem.

Recurring revenue
Hardware sale plus SaaS, not hardware alone

A one-time hardware margin at installation, plus a recurring per-device SaaS revenue line that compounds with every unit shipped.

Fleet visibility
Predictive maintenance across your installed base

Fleet dashboard and CMMS integration turn a warranty return — the first signal something went wrong, weeks late — into a scheduled service call instead of an emergency one.

Dual-sensor architecture

Inlet vs outlet. The only measurement that matters.

TDSBot uses a two-sensor configuration: one sensor on the feed water line before the membrane, one on the permeate output. The reject ratio — (TDS_inlet − TDS_outlet) / TDS_inlet — gives a continuous, real-time view of membrane performance that a single outlet sensor cannot provide. A high outlet TDS reading alone doesn't tell you whether the membrane is degraded or the inlet water has changed. The ratio does.

Membrane degradation detectionReject ratio falls as membrane ages, fouling increases, or physical damage occurs.
Inlet water change detectionTDS_inlet changes without affecting reject ratio — no false alarms from seasonal source water variation.
Bypass valve drift detectionTDS_blend rises while TDS_membrane stays stable — identifies valve failure independent of membrane condition.
Service triggerAlert fires when reject ratio drops below your configurable threshold — before the customer notices output quality change.
Data for warranty decisionsContinuous log of inlet quality and reject ratio — distinguish membrane failure from poor inlet water quality.
RO system architecture · standard configuration
Feed waterS1 · TDS_inlet420 ppm
RO membrane95% rejection—
Permeate outS2 · TDS_outlet21 ppm
Reject / concentrateHigh TDS drain—
Reject ratio formula
RR = (TDS_inlet − TDS_outlet) ÷ TDS_inlet × 100
Bypass valve drift detection

Two failure modes. One looks identical without dual sensors.

In RO systems with a blend valve — which mixes permeate with bypass water to raise mineral content — there are two separate failure modes that both cause outlet TDS to rise. Only a dual-sensor configuration can tell them apart.

Membrane failure
Both TDS readings rise together
TDS_inlet410 ppm
TDS_membrane185 ppm ↑
TDS_blend142 ppm ↑
Reject ratio: 55% — well below threshold. Membrane is degraded. Service action: replace membrane. Warranty claim: membrane failure.
Bypass valve drift
Only blend TDS rises — membrane is fine
TDS_inlet412 ppm
TDS_membrane22 ppm ✓
TDS_blend138 ppm ↑
Membrane reject ratio: 94.7% — healthy. Blend ratio has drifted. Service action: recalibrate or replace blend valve. No membrane warranty claim justified.
Membrane degradation
TDS_membrane rises AND TDS_blend rises. Both sensors track each other. Rejection is falling.
Bypass valve drift
TDS_membrane stays flat. TDS_blend rises. The valve is allowing too much raw water through — the membrane is fine.
Why it matters for manufacturers
A warranty claim for "poor water quality" can be instantly investigated — membrane fault vs valve fault vs inlet water change. Three different resolutions, distinguishable in seconds from sensor data.
Deployment configurations

Commercial RO, residential RO,
and point-of-use filtration

TDSBot dual-sensor architecture adapts to three distinct product categories. Each has different monitoring priorities, different customer expectations, and different service economics.

Commercial RO
Commercial RO systems · institutional and food service

Large-format RO systems serving hospital kitchens, food manufacturers, commercial laundry, and large food service operations. Water volumes are high, membrane replacement costs are significant, and downtime is unacceptable. The reject ratio alert fires before quality drops — creating a scheduled service call rather than an emergency replacement.

TDSBot configuration
S1 on feed line · S2 on permeate · alert at configurable RR threshold · CMMS integration for automatic work order creation · monthly report for HACCP supporting documentation
Residential RO
Under-sink and whole-home RO systems

Home RO systems with a blend valve to restore minerals to permeate output. The homeowner has no way to know when their membrane degrades — and no baseline to compare to. TDSBot embeds in the system at manufacture, connects to a consumer app, and provides the first consumer-facing membrane health indicator in the market.

TDSBot configuration
S1 on feed · S2 on blend output · consumer mobile app · white-label under your brand · monthly health digest · pre-order membrane reminder with 30-day lead
Point-of-Use Filtration
Inline filters, countertop RO, under-sink systems

For non-RO inline filters (activated carbon, sediment, ion exchange), TDSBot uses a single-sensor configuration on the outlet — tracking TDS drift against the onboarded baseline as the primary filter depletion signal. For countertop RO units, dual-sensor configuration applies with the same reject ratio model.

TDSBot configuration
Single S1 on outlet (non-RO) · dual S1/S2 for RO · filter depletion model using volume flow × TDS drift · alert 30–50 days before exhaustion
OEM commercial model

Partner with TDSBot.
One platform underneath.

The WaterTDS platform handles the hard parts — AWS IoT Core connectivity, data pipeline, alert logic, API, and security — so you don't have to build any of it.

Co-branded

TDSBot hardware ships with co-branding (e.g. "Powered by TDSBot"). Your customers use the WaterTDS app with your property details. You earn revenue share on SaaS subscriptions. Fast to market, no development overhead on your side.

HardwareTDSBot branded with your co-brand sticker
SaaSWaterTDS platform, customer creates account
RevenueYou earn % of SaaS subscription per device
PricingHardware at dealer cost + SaaS revenue share
Time to market2–4 weeks
Why partner instead of building it yourself
What you'd have to build vs. what's already built
Build internallyWaterTDS
Cloud infrastructure (AWS IoT Core, data pipeline)Months of engineeringIncluded
Degradation models, alert logicNeeds field data you don't have yetIncluded
API for your own appBuild from scratchIncluded
Connectivity security (MQTT/TLS 1.3)Build from scratchIncluded
Co-branded dashboard + appBuild from scratchIncluded
Time to market—2–4 weeks

Have something else in mind — fully embedded white-label, direct API access, a different revenue structure?

Pricing

OEM pricing is negotiated.
Not listed.

End-customer plan pricing, OEM hardware cost, white-label fees, and wholesale SaaS rates are all established at contract stage based on volume commitments and integration model. We don't publish a price list for OEM partnerships — the right number depends on your product, your channel, and your scale.

Traditional RO
Sell purifier——
Done
Connected RO, with TDSBot
Sell purifierRecurring SaaS monitoringFleet dashboard + CMMS integration
Compounding revenue, every unit shipped
Hardware
OEM unit cost

Tiered by annual volume commitment. Significant savings at scale. White-label manufacturing available.

SaaS monitoring
OEM wholesale rate

Recurring per-device revenue. OEM wholesale rate established at contract stage. Compounds with every unit shipped.

White-label
Platform fee

Covers dashboard customisation, firmware branding, API access, and ongoing platform development.

Regulatory case study · Japan 水道法 · PFAS

A real regulatory tailwind. Here's exactly what we can claim.

Japan's Water Supply Act added PFOS and PFOA as mandatory water quality standards in April 2026 — the first time these contaminants have been statutory rather than advisory, with quarterly lab testing now required. Our active MOU partner is a Japanese RO manufacturer deploying TDSBot as the monitoring layer between those mandatory tests.

Active MOU — Japanese RO manufacturer · deployment in progress
50 ng/L
PFOS+PFOA combined statutory limit
4×/yr
Mandatory minimum test frequency
365d
Continuous reject ratio coverage between tests
水道法
Water Supply Act — April 2026 施行
Monitors membrane rejection rate in real time

Continuous reject ratio from dual TDS sensors. When the ratio is high, the membrane is performing. When it falls, performance is declining across all contaminants the membrane removes by the same mechanism.

Can do · continuous monitoring
Infers PFAS removal performance from reject ratio

RO membranes remove PFAS and TDS by the same physical mechanisms (size exclusion and charge repulsion). A declining reject ratio indicates declining removal efficiency across the board — including PFAS. This is a reasonable inference for operational purposes, not a measured result.

Inference · reasonable proxy
Supports compliance documentation between lab tests

Continuous reject ratio records show the membrane was performing throughout the quarter. TDSBot data supports — but does not replace — the mandatory quarterly lab test.

Can do · supporting documentation
Generates statutory PFAS test results

TDSBot does not measure PFAS concentration. It measures TDS. Statutory PFAS testing (PFOS/PFOA ≤50 ng/L under Japan 水道法, or EPA health advisory limits) requires laboratory analysis of a water sample by accredited methods. TDSBot cannot substitute for this requirement.

Cannot do · lab testing required
OEM partner perspectives

From product managers and
technical directors

“The bypass valve drift detection was the feature that clinched our decision. We were getting warranty claims that were actually valve failures — not membrane failures. Our team was replacing membranes unnecessarily. TDSBot's dual-sensor data identified the real fault in every case. We've eliminated a whole category of warranty cost.”
HMHiroshi MatsudaHead of Product · Japanese RO manufacturer · MOU partner
“Since deploying TDSBot monitoring, customers call our service line differently. Instead of "my water tastes wrong," they say "my reject ratio dropped to 79% on Tuesday." That diagnostic specificity cuts average service call time by 35 minutes. The data pays for itself on the first call.”
KPKevin PressleyTechnical Director · Commercial RO manufacturer · OEM pilot
“The bypass valve drift detection was the feature that clinched our decision. We were getting warranty claims that were actually valve failures — not membrane failures. Our team was replacing membranes unnecessarily. TDSBot's dual-sensor data identified the real fault in every case. We've eliminated a whole category of warranty cost.”
HMHiroshi MatsudaHead of Product · Japanese RO manufacturer · MOU partner
“Since deploying TDSBot monitoring, customers call our service line differently. Instead of "my water tastes wrong," they say "my reject ratio dropped to 79% on Tuesday." That diagnostic specificity cuts average service call time by 35 minutes. The data pays for itself on the first call.”
KPKevin PressleyTechnical Director · Commercial RO manufacturer · OEM pilot
Technical & commercial questions

From product managers and
integration engineers

TDSBot uses a dual-sensor architecture: one sensor on the inlet (before the membrane) and one on the outlet (permeate side). The reject ratio — (TDS_inlet − TDS_outlet) / TDS_inlet × 100% — indicates membrane performance. A healthy RO membrane achieves 90–98% rejection. As the membrane degrades, reject ratio falls and outlet TDS rises. TDSBot tracks reject ratio continuously and alerts when it drops below a configurable threshold, providing early warning of membrane degradation before it becomes visible to the end customer.

Yes. For RO systems with a blend valve, TDSBot deploys three measurement points: TDS_inlet (before membrane), TDS_in_membrane (pure permeate, after membrane, before blend valve), and TDS_in_blend (after blend valve, final output). The membrane reject ratio is calculated from TDS_inlet and TDS_in_membrane — the true membrane performance metric, unaffected by blend valve setting. TDS_in_blend is tracked separately: when the blend valve drifts open, TDS_in_blend rises while TDS_in_membrane remains stable, distinguishing valve drift from membrane degradation. The discriminator Δ = TDS_in_blend − TDS_in_membrane tracks whether the blend ratio is stable. This requires a three-sensor deployment; the third sensor is available as an optional add-on.

No — TDSBot does not directly measure PFAS or any specific contaminant. For RO systems, TDSBot monitors the reject ratio (inlet TDS vs outlet TDS). Because RO membranes remove TDS and PFAS by the same physical mechanism (size exclusion and charge repulsion), a declining reject ratio indicates that the membrane's overall removal efficiency is declining — which includes PFAS removal. This makes reject ratio monitoring a useful proxy for membrane performance between mandatory laboratory PFAS tests. TDSBot does not generate statutory PFAS test results.

TDSBot uses quick-connect fittings compatible with 1/4″ to 1 1/2″ water lines. In a standard RO installation: Sensor 1 (TDS_inlet) is installed on the feed water line before the membrane housing. Sensor 2 (TDS_outlet) is installed on the permeate line after the membrane — or on the blend output for systems with a blend valve. The ESP32-S controller connects to WiFi during commissioning and communicates with AWS IoT Core over MQTT/TLS 1.3. Installation is 15–20 minutes per system.

This depends on your membrane specification and the end application. For food service or medical applications, a threshold of 88–90% (i.e. alert fires when rejection drops below 88%) is typical. For residential or general commercial use, 82–85% is common. The threshold is configurable per device through the API or dashboard at onboarding. A new membrane typically achieves 94–98% rejection; as it ages, rejection gradually falls. We recommend setting the threshold 5–8 percentage points above the application's minimum acceptable rejection rate.

Yes. WaterTDS's current OEM model is co-branded: TDSBot hardware ships with your co-branding (e.g. “Powered by TDSBot”), your customers use the WaterTDS app with your product details, and you earn revenue share on SaaS subscriptions. Raw sensor readings always belong to you and your customer; alert logic and analytics models are WaterTDS platform IP delivered as outputs. Typical configuration takes 2–4 weeks. Fully embedded white-label, direct API access, or a different revenue structure — we're not set up for those as standard offerings today, but talk to us if that's what you need.

OEM hardware pricing is negotiated based on annual volume commitment. Indicative tiers: 500–2,000 units annually — modest discount on list prices; with tiered discount on larger quantities. Contact support@watertds.com to begin the conversation.
Ready to make your RO system intelligent?

Embed membrane health monitoring
in your next product generation

Active MOU and pilot deployments with leading RO manufacturers across commercial, residential, and institutional segments. White-label hardware, co-brand, or additional architectures discussed under NDA.

Cookies. Some keep the site working. Analytics run only if you allow them, and we use no advertising cookies. Change your answer any time in the Cookie Policy.

WaterTDS Inc
Inline IoT water quality monitoring for coffee, food service, water treatment, facilities and home filtration.
MQTT over TLS 1.3 · AWS IoT Core · ESP32-S
Disclaimers: TDSBot measures TDS, conductivity, temperature and flow rate directly. LSI, Larson-Skold Index, hardness and alkalinity estimates are derived from onboarded utility water chemistry data — not directly measured inline. TDSBot does not measure PFAS, bacteria, or other biological or chemical contaminants; for RO systems, reject ratio is used as a proxy indicator of membrane performance only. TDSBot monitoring data is not a substitute for accredited laboratory testing or statutory water quality compliance testing. All third-party brand names are trademarks of their respective owners; WaterTDS Inc. claims no affiliation, endorsement or partnership unless separately announced.

© 2020 - 2026 Water TDS Inc.  All rights reserved.

TDSBot® · PWSID cross-referenced · EPA SDWIS · SCA Water Standards