If you’re dealing with high-TDS wastewater, the short answer is this: DTRO (disc-tube reverse osmosis) usually wins when your feed is dirty, scaling-prone, and loaded with suspended solids, while conventional RO is the better pick when your stream is clean, consistent, and you want the lowest cost per cubic meter of permeate. I’ve commissioned both types on landfill leachate, plating rinse water, and chemical plant effluent, and the deciding factor is almost never the membrane chemistry itself. It’s how much pretreatment the feed demands and how often you can afford to stop the plant for cleaning.
Let me walk you through how I actually make this call on real projects, because the brochure comparison sheets leave out the parts that cost you money.
What “High TDS” Actually Means in Practice
People throw around “high TDS” like it’s one condition. It isn’t. From a design standpoint, I break it into three bands:
- 3,000–15,000 mg/L TDS: Brackish territory. Standard RO handles this comfortably at 15–25 bar, provided pretreatment is solid.
- 15,000–35,000 mg/L TDS: Approaching seawater salinity. You’re now in high-pressure RO territory, and scaling control becomes the whole game.
- Above 35,000 mg/L TDS: Conventional RO gets expensive fast. This is where DTRO, and often a hybrid with evaporation, starts to make sense.
But TDS alone doesn’t tell you which membrane to use. I’ve seen 8,000 mg/L leachate destroy a conventional RO train in six weeks because of colloidal silica and organic fouling. I’ve also seen 40,000 mg/L brine run on high-pressure RO for years because the feed was clean and well-conditioned.
The real question is: what else is in the water? Suspended solids, hardness, silica, organics, oil, and heavy metals all matter more than the TDS number by itself.
The Core Mechanical Difference Between DTRO and RO
Conventional RO uses spiral-wound elements. Flat membrane sheets are wrapped around a central permeate tube with a feed spacer in between. Feed flows axially through the spacer channel, which is typically 0.7–1.0 mm thick. That narrow channel is the weak point. Plug it with solids and you lose the element.
DTRO uses disc-shaped membrane stacks mounted on a central rod, compressed hydraulically inside a pressure vessel. Feed enters through a central channel and flows radially across the disc surface. The open channel design is much more forgiving of particulates, and you can open the stack and replace individual discs instead of scrapping an entire element.
That mechanical difference drives everything else — pretreatment requirements, cleaning frequency, recovery, and cost.
Side-by-Side Comparison
| Параметр | Conventional RO (spiral-wound) | DTRO (disc-tube) |
|---|---|---|
| Typical feed TDS range | Up to ~35,000 mg/L with high-pressure elements | Up to ~60,000–80,000 mg/L in staged configurations (project-dependent) |
| Suspended solids tolerance | Very low; SDI < 3–5 required | Higher; can tolerate SDI 5–10 with proper flushing |
| Pretreatment demand | Extensive — DAF, media filtration, cartridge filters, antiscalant | Moderate — coarse filtration and pH adjustment often sufficient |
| Fouling/scaling resistance | Lower; narrow spacer channel clogs easily | Higher; open channel and turbulent radial flow resist deposits |
| Cleaning frequency | Weekly to monthly for dirty feeds | Monthly to quarterly for comparable feeds |
| Recovery per stage | 75–85% typical for brackish | 70–80% per stage; multi-stage raises overall recovery |
| Рабочее давление | 15–25 bar brackish; 55–70 bar seawater-class | 40–90 bar depending on concentration stage |
| Membrane replacement | Replace whole element | Replace individual discs |
| Площадь занимаемой территории | Компактный | Larger per unit permeate |
| Капитальные затраты | Lower | Higher — often 1.5–2.5× for equivalent capacity |
| Best fit | Clean, consistent, well-pretreated feeds | Dirty, variable, scaling-prone high-TDS streams |
Those numbers are indicative design ranges, not guarantees. Every project shifts them based on water chemistry and operating philosophy.
Pretreatment: Where the Real Cost Lives
Here’s what I tell every client who asks me to compare membrane options: the membrane is maybe 20% of your total cost of ownership. Pretreatment, cleaning chemicals, downtime, and labor eat the rest.
A conventional RO system on high-TDS wastewater typically needs:
- Equalization and pH adjustment
- Coagulation/flocculation and clarification or DAF
- Media filtration and often ultrafiltration
- Cartridge filtration at 5 µm as a final guard
- Antiscalant and biocide dosing
A DTRO membrane system can often skip the ultrafiltration stage and run with coarser prefiltration. That’s a meaningful capex and opex saving, and it’s why DTRO shows up so often on leachate and industrial brines where the feed quality swings day to day.
If you’re designing toward zero liquid discharge, the membrane is only one piece. You’ll still need a concentrator downstream — usually an Испаритель MVR or a multi-effect train — to take the RO reject from 15–20% solids up to the crystallizer feed. I’ve written about the full ZLD system architecture elsewhere if you want the complete picture.
When DTRO Clearly Wins
I reach for DTRO in these situations without hesitation:
1. Landfill leachate
Leachate is the poster child for DTRO. It combines high TDS, high ammonia, humic and fulvic organics, heavy metals, and unpredictable composition as the landfill ages. Spiral-wound membranes foul within weeks. DTRO stacks handle it with a cleaning cycle measured in months, not days.
2. High-scaling brines
If your feed sits near or above saturation for calcium sulfate, barium sulfate, or silica, the open DTRO channel tolerates supersaturation better and lets you push recovery higher before scaling forces a shutdown.
3. Variable feed quality
Plants that batch-process different products — specialty chemicals, pharmaceutical intermediates, metal finishing — see feed composition swing weekly. DTRO’s fouling tolerance gives you operating margin that spiral-wound elements don’t.
4. Tight footprint isn’t a constraint
DTRO takes more floor space per unit of permeate. If you have the room, that’s a non-issue. If you’re retrofitting into an existing building, it can be the deciding factor against DTRO.
When Conventional RO Is the Right Call
Don’t over-engineer. I specify conventional RO when:
- The feed is already well-treated — for example, RO polishing after biological treatment and ultrafiltration
- TDS is below 15,000 mg/L and stable
- You need the lowest possible capex and the site has room for standard skids
- Power cost is high and you want to minimize high-pressure pumping
- Permeate quality targets are moderate and you don’t need the extra fouling margin
On a clean brackish stream, a high-pressure RO membrane train will deliver better energy efficiency and lower maintenance than DTRO every time. Paying for fouling tolerance you don’t need is just wasted capital.
Energy and Lifecycle Cost Reality Check
Specific energy consumption for RO is roughly proportional to feed pressure and recovery. On a brackish feed at 20 bar, expect 1.5–3 kWh per m³ of permeate. Push to 60–70 bar for high-TDS feed and you’re looking at 4–8 kWh per m³ before you account for pretreatment pumping and dosing.
DTRO operates at similar or slightly higher pressures than high-pressure RO for the same duty, but the trade-off is cleaning frequency and membrane life. In my experience on leachate projects, DTRO membrane disc replacement intervals run 3–5 years, while spiral-wound elements on the same feed might need replacement every 12–18 months. That difference often pays back the higher DTRO capex within the first replacement cycle.
If you need to concentrate the reject stream further, evaporation takes over. A испаритель с механическим сжатием пара (MVC) typically consumes 15–40 kWh per m³ of distillate depending on boiling point elevation and compressor efficiency — much higher than RO, which is why you want the membrane to do as much of the concentration work as possible.
Rule of thumb I use: push membrane concentration until osmotic pressure or scaling risk forces you to stop, then hand off to evaporation. Every cubic meter you remove with membranes instead of evaporators saves you roughly 10–30 kWh.
Maintenance and Failure Patterns
What actually breaks in the field:
Conventional RO
- Spacer channel plugging — the number one killer on dirty feeds. Once flow channels block, cleaning rarely restores full performance.
- Biofouling — biofilm forms in the feed spacer and is nearly impossible to remove completely.
- Scaling at the tail end — the last elements see the highest concentration factor and scale first.
- O-ring and glue line failures — usually from pressure cycling or thermal stress.
DTRO
- Seal and gasket wear — the hydraulic compression system needs periodic seal replacement.
- Disc cracking — usually from over-torque during reassembly or pressure spikes.
- Concentrate-side scaling — still happens, but cleaning is easier because you can open the stack.
- Higher pump maintenance — high-pressure pumps run harder and need more attention.
Neither technology is maintenance-free. The question is whether your team can handle the maintenance in-house or whether you need to fly in a service technician every time something fails. That logistics cost is real and often underestimated.
How I Structure the Selection Decision
When a client sends me a water analysis and asks which membrane to use, I run through this sequence:
- Characterize the feed properly. Not just TDS. I want SDI, turbidity, hardness, silica, iron, manganese, organics (TOC/COD), oil and grease, and pH range over time.
- Define the permeate quality target. If you need <500 mg/L TDS permeate, that’s a different design than <2,000 mg/L.
- Estimate the required recovery. High recovery concentrates everything and raises scaling risk. This often decides the technology.
- Check the pretreatment you can realistically operate. A perfect pretreatment train that nobody maintains is worse than a forgiving membrane.
- Look at the downstream. If you’re going to ZLD, the membrane reject composition drives evaporator and crystallizer design.
- Run the lifecycle cost. Include membrane replacement, cleaning chemicals, labor, energy, and downtime. Not just capex.
I’ve seen projects where the “cheaper” RO option cost 40% more over ten years than the DTRO alternative because of downtime and membrane replacement. I’ve also seen the opposite. The math has to be run on your specific water.
Hybrid Configurations Worth Considering
You don’t always have to pick one. Some of the most robust designs I’ve built use both:
- DTRO first stage, RO polishing second stage. DTRO handles the dirty, high-TDS feed and produces intermediate-quality permeate. Conventional RO polishes it to final spec at lower pressure. This cuts total energy and extends RO element life.
- RO concentration followed by DTRO on the reject. Standard RO gets you to 70–75% recovery cheaply. DTRO squeezes another 10–15% out of the reject before evaporation takes over. This reduces evaporator size and energy significantly.
- Membrane + MVR evaporator. The membrane does the bulk concentration; the evaporator finishes the job and produces distillate for reuse. This is the backbone of most modern ZLD plants.
For leachate specifically, I’ve had good results with a DTRO two-stage design feeding a многоступенчатый выпариватель for final concentration. The evaporator handles the last 20% of volume, which is where the energy cost would otherwise explode.
Common Mistakes I See on High-TDS Projects
After enough commissioning trips, the failure patterns repeat:
- Designing on a single water analysis. High-TDS industrial streams vary. I always ask for at least a 12-month historical range, or run a pilot if the data doesn’t exist.
- Underestimating pretreatment. The membrane vendor says “SDI < 5” and the client builds the cheapest possible filtration. Six months later the elements are plugged.
- Ignoring osmotic pressure at high recovery. At 80% recovery on 30,000 mg/L feed, the concentrate side sees 150,000 mg/L. That’s 100+ bar of osmotic pressure. Standard elements can’t handle it.
- No cleaning-in-place system. If you’re running high-TDS feed without a proper CIP skid, you’re planning to fail.
- Forgetting the concentrate. Every membrane produces a reject stream. If you haven’t designed the disposal or further concentration step, the project isn’t finished.
Часто задаваемые вопросы
Can DTRO handle higher TDS than conventional RO?
Yes, in practice. DTRO systems are commonly applied to feeds in the 30,000–60,000 mg/L range, and staged configurations can go higher. Conventional RO can technically reach similar concentrations with seawater-class elements, but fouling and scaling risk rises sharply, and recovery has to be kept low. The practical limit for spiral-wound RO on dirty high-TDS streams is usually much lower than the theoretical one.
Which membrane has lower operating cost?
It depends on feed quality. On clean feeds, conventional RO wins on energy and membrane replacement cost. On dirty or scaling-prone feeds, DTRO often wins overall because cleaning frequency drops, membrane life extends, and downtime shrinks. I’ve seen DTRO deliver 20–30% lower total operating cost on leachate, and conventional RO deliver 30–40% lower cost on clean brackish water. There’s no universal answer.
How often do DTRO membranes need cleaning?
On typical landfill leachate, I expect a cleaning cycle every 4–12 weeks. On cleaner industrial streams, quarterly cleaning is common. Conventional RO on the same leachate would need weekly cleaning, sometimes more often. The difference comes from the open channel design and the ability to flush the stack thoroughly between cycles.
What pretreatment does DTRO require?
Less than conventional RO, but not zero. At minimum, I specify coarse screening or strainers, pH adjustment if needed, and often a media filter or bag filter. Some designs add antiscalant dosing. Ultrafiltration is usually optional for DTRO but mandatory for conventional RO on the same feed.
Is DTRO worth the higher capital cost?
On high-TDS, fouling-prone streams, yes — usually within 3–5 years through reduced membrane replacement and downtime. On clean feeds, no. Run the lifecycle cost with your actual water data before deciding. If the payback is longer than your project horizon, conventional RO is the better choice.
Can I retrofit DTRO into an existing RO plant?
Sometimes. If the existing high-pressure pumps and piping can handle the DTRO pressure and flow requirements, and you have floor space for the larger skid, a retrofit is feasible. More often, I recommend a hybrid approach: keep the existing RO for polishing and add a DTRO stage upstream or on the reject. This uses the existing assets and adds fouling tolerance where it’s needed.
What happens to the DTRO concentrate?
It goes to further concentration or disposal. On ZLD projects, the DTRO reject feeds an evaporator and crystallizer. On discharge projects, it goes to a permitted outfall or a brine disposal contractor. The concentrate volume and composition drive the downstream design, so this has to be planned from day one — not after the membrane system is running.
Bottom Line
There’s no universal winner between DTRO and RO on high-TDS wastewater. The right choice comes down to three things: how dirty your feed is, how much pretreatment you can realistically operate, and what your lifecycle cost looks like with your actual water chemistry.
My default position: if the feed is clean and stable, use conventional RO and save the capital. If the feed is dirty, scaling-prone, or variable — leachate, plating waste, chemical brines — DTRO will usually pay for itself through uptime and membrane life. And if you’re heading toward zero liquid discharge, plan the membrane and evaporator stages together from the start, because the membrane’s job is to make the evaporator’s job as small as possible.
Get the water analysis right, run the numbers honestly, and the decision usually makes itself.