Almost everyone buys these two things in the wrong order.
They buy the automatic litter box first — it is the exciting purchase, it costs between S$280 and S$1,700 in Singapore, and it is the thing that gets researched. Then they pour in whatever litter they already had. Then, a week later, the machine starts throwing away clean litter, or the clumps smear, or the rake jams, and they conclude the box is faulty.
The box is usually fine. In Singapore the two default choices are incompatible with each other by default, and nobody warns you, because the litter and the machine are sold by different people.
This article is the order we think a first-time buyer should work in, and the five numbers that decide it. We sell both the litter and the box, so read the last section before you decide anything — it is where we say what we cannot tell you.
The Singapore trap, in two numbers
The default cat litter in Singapore is tofu litter. It dominates what actually sells here — on the marketplace where most of this country's pet supplies move, most of the best-selling litters are tofu.
Standard tofu litter is extruded as sticks around 3 mm across.
The grates in sifting automatic boxes are built for pellets around 1.5–2.0 mm.
That is the whole problem. A 3 mm stick does not fall through cleanly. It lodges. The machine then reads the blocked grate as waste and drops clean litter into the bin, and you refill the box twice as often while thinking the machine is broken.
We have written that up in detail in Automatic Litter Boxes and Tofu Litter. What follows is what to do before you get to that point.
Five numbers decide whether a litter works in a machine — and they are all in one standard
There is a written standard for tofu cat litter: T/CIQA 11-2020, issued by the China Entry-Exit Inspection and Quarantine Association. It grades litter as Premium, Grade 1 or Qualified across thirteen parameters. Five of those parameters are the ones that decide how a litter behaves inside a moving machine, and almost no seller quotes any of them.
| Parameter | Premium | Grade 1 | Qualified | What it does inside a machine |
|---|---|---|---|---|
| Clump strength | ≥90% | ≥80% | ≥75% | Whether the clump survives being lifted, tipped and dropped. A weak clump breaks apart mid-cycle and smears through the bed. |
| Disintegration time | ≤30 s | ≤50 s | ≤60 s | How fast a clump falls apart in water. Relevant if you flush; also a proxy for how much handling a clump tolerates. |
| Hardness | ≥35 N | ≥35 N | ≥35 N | Whether pellets survive being pushed by a rake instead of being crushed into powder. |
| Powder content | ≤0.5% | ≤1% | ≤1.5% | How much dust is already in the bag before the machine touches it. |
| Pulverisation rate | ≤0.5% | ≤0.8% | ≤1% | How much more dust the litter makes when it is abraded — which is exactly what a rake does to it, several times a day. |
The last one is the one people miss. An automatic box is a machine that abrades your litter on a schedule. A litter that is fine in a tray, where nothing moves, can turn to powder in a sifter. Powder content tells you what is in the bag; pulverisation rate tells you what happens after the machine has had it for a month.
The standard also specifies bulk density at 550 ± 50 g/L, which does not affect performance but decides how much litter one fill costs you. We come to that next.
How much litter does one fill actually take?
Here is the arithmetic almost nobody does before buying a machine.
Our own box has a litter area of 38.5 × 44 cm, and needs a bed of at least 5 cm — below that the mechanism scrapes the tray and the level sensor mis-reads. That is:
38.5 × 44 × 5 cm = 8.47 litres of litter, per fill.
Turn that into bags, and the answer depends on the litter's bulk density:
| If the litter's bulk density is… | …one 8.47 L fill weighs | …in 2.1 kg bags |
|---|---|---|
| 600 g/L (top of the standard's band) | 5.08 kg | 2.4 bags |
| 550 g/L (centre of the band) | 4.66 kg | 2.2 bags |
| 500 g/L (bottom of the band) | 4.24 kg | 2.0 bags |
| 417 g/L (typical of retail "6 L" packs) | 3.53 kg | 1.7 bags |
So the first fill of a machine takes roughly two bags of litter, not one. After that you are topping up, not refilling, so the ongoing cost is lower — but the first shop is bigger than people expect, and it is worth knowing before the box arrives and you have one bag in the cupboard.
It also means something less obvious: a denser litter costs more per fill and lasts no longer, because the machine consumes litter by volume, not by weight. If you are comparing two litters and one is denser, the denser one is worse value in a machine at the same price per kilogram. This is the opposite of the intuition most people have.
Now ours, graded honestly
We publish three of the thirteen T/CIQA 11-2020 parameters for our tofu litter, from third-party testing, with report numbers you can verify:
| Parameter | Ours | Which grade that meets |
|---|---|---|
| Clump strength | 94.1% | Premium (≥90%) |
| Pulverisation rate | 0.36% | Premium (≤0.5%) |
| Powder content | 0.66% | Grade 1 (≤1%), not Premium |
Under the standard's own grading the lowest parameter governs. So on the parameters we have measured, our litter is Grade 1, not Premium. We would rather say that plainly than quote the two flattering numbers and leave the third out.
And the rest of the table we have simply not tested: disintegration time, hardness, water absorption, clump weight, and bulk density are all unmeasured on our side. Hardness is the one that bothers us most in this context, because it is the parameter that predicts how a pellet holds up under a rake, and we sell a machine with a rake.
Pellet size we do have: 1.5 mm, which is inside the 1.5–2.0 mm range that sifting boxes are built for. That is the specification that matters most for the pairing, and it is the reason our litter works in a machine when standard 3 mm tofu does not.
Bulk density we are currently measuring ourselves and will publish with the method written out. Until then, the two-bags-per-fill figure above is a range, not a number.
The one about the box, which is worse
The automatic box we sell does not have a published minimum detection weight — the weight below which the machine does not register that a cat is inside. Litter-Robot publishes theirs. We do not have ours, we have asked for it, and until we have it that row on our spec sheet stays blank.
If that number is what decides your purchase, it should decide it in someone else's favour. We would rather you bought a machine whose safety floor you can look up.
Before any of this: you may not need a machine
No automatic litter box is a kitten product. The published minimum detection weights across the category run from about 0.9 kg to 1.4 kg, and most kittens at 8–12 weeks weigh less than that. One major manufacturer's own guidance says 12 weeks and up, and explicitly not under 8 weeks. If you have a kitten, buy a cheap open tray and buy the machine when the cat is grown.
And for a first-time owner with one adult cat and a normal schedule, we will say the unhelpful thing: an ordinary tray plus a litter that suits it beats a cheap automatic box plus the wrong litter, every time, and it costs about S$15.
An automatic box earns its money when you have more than one cat, when you travel, when scooping is physically difficult, or when the box lives somewhere you would rather not visit twice a day. If none of those is true for you, the machine is a convenience purchase, and there is nothing wrong with that — but it is worth knowing which kind of purchase you are making.
The order we would work in
- Decide whether you need a machine at all. One adult cat, home most days, box somewhere convenient — a tray is fine.
- Measure the space and find the power point before you shortlist anything. Service yards in older HDB flats often have exactly one socket, and the washing machine is in it.
- Choose the litter design first: sifting grate, rake, or rotating drum. This decides what litter you can use for as long as you own the machine.
- Now check the litter. For a sifter: pellet size 1.5–2.0 mm. Crystal and non-clumping litters are generally incompatible with rake designs.
- Ask both sellers for numbers, not adjectives. For the litter: clump strength, powder content, pulverisation rate, and which standard they were measured against. For the box: minimum detection weight. "Ultra low dust" with no figure and no method is an adjective.
- Buy two bags of litter for the first fill, not one.
- Set the cycle delay to 10–15 minutes and fill to at least 5 cm, then recalibrate the level sensor whenever you change litter type.
- Open the app monthly and apply firmware updates. Free, and the highest-value safety habit in this category.
If you want to go deeper: Automatic Litter Boxes and Tofu Litter covers the specifications and the settings, 6 Litres of What? covers why you cannot compare litter prices here, Cat Litter Dust covers the two dust numbers and what they mean, and Tofu Cat Litter in Singapore is the general buyer's guide. Everything we have had tested, and a plain list of what we have not, is on The numbers we publish.
Written September 2026. T/CIQA 11-2020 (豆腐猫砂规范, Specification on tofu cat litter) is a group standard of the China Entry-Exit Inspection and Quarantine Association, effective 1 January 2021; the grade thresholds quoted are from its Table 1. Our own litter figures come from third-party testing under that standard, with report numbers published on this site for verification; disintegration time, hardness, water absorption, clump weight and bulk density are not among the parameters we have had tested. Litter box dimensions are our own product's published specifications. Minimum detection weights quoted as a category range are from Litter-Robot and PetSafe published specifications; the kitten age guidance is Neakasa's published guidance. Singapore automatic litter box prices are retailer listings in September 2026 and change frequently.
