Four physical tests decide whether a batch is released. This is what each one is really asking, which apparatus answers it, and the parts of the bench that quietly ruin results when nobody is watching.
Bangladesh makes most of its own medicines and exports a good deal of what it makes. The physical testing bench in a tablet plant is therefore not a formality. It is the part of the quality control laboratory an auditor walks to first, because four inexpensive tests decide whether a batch is released and because those four tests are easy to run badly without anyone noticing for months.
The tests are dissolution, disintegration, friability and hardness. They look similar on a purchase order and they answer completely different questions.
Read in that order, hardness is an early warning, friability is a handling check, disintegration is a screen and dissolution is the release test. A laboratory that owns only a dissolution tester finds out about a problem at the end. A laboratory that owns a hardness tester finds out during compression, when the batch can still be saved.
The pharmacopoeias define numbered apparatus, and the number that appears in the product monograph is not negotiable. Buying the wrong one is the most expensive mistake available on this bench.
| Apparatus | Form | Typical use |
|---|---|---|
| USP 1 and 2 | Basket and paddle in a 1000 ml vessel | Immediate release tablets and capsules. The overwhelming majority of routine work |
| USP 3 | Reciprocating cylinder | Extended and sustained release forms, and any product where the medium has to change during the run |
| USP 4 | Flow through cell | Poorly soluble products, low dose sustained release, implants and forms where sink conditions cannot be held in a vessel |
| USP 5 and 6 | Paddle over disc and rotating cylinder | Transdermal systems |
Erweka cover these as distinct machines rather than as options on one frame. The Digital Dissolution Tester DT 950 Series (DT 956, DT 957, DT 958) is the workhorse: an embedded PC, a seven inch touch display, an integrated audit trail, and a TestAssist routine that walks the operator through configuration so only compliant parameters can be entered. It upgrades from six to eight test stations, takes automatic tablet drop and per vessel temperature sensors, and switches between low head and high head modes. The Digital Dissolution Tester DT 9510 Series (DT 9512, DT 9513, DT 9514) is the same architecture with up to 14 stations driven from one motor and a large water bath whose thermal inertia holds temperature steady, so two batches run in parallel under identical conditions.
Where budget is the constraint and the work is straightforward, the Dissolution Tester DT Light Series (DT 126 Light, DT 128 Light) covers USP methods 1, 2, 5 and 6 with six or eight stations on a fixed high head drive, a one piece moulded PET water bath and an external flow through heater that keeps pump vibration away from the vessels. It is a real instrument with fewer features, not a cheap imitation of one.
For extended release work the Dissolution Tester BioDis RRT 10 (USP 3 / USP 7) is the right machine, because it moves the sample between rows rather than moving the medium, so a different medium in each row reproduces the pH sequence of the gastrointestinal tract. And for products that will not hold sink conditions in a litre of medium, the Flow-Through Cell DFZ II (USP 4) runs medium past the sample continuously, with a standardised cell head that fits every cell body on offer and each cell heated individually on a rotary switch.
Ask a laboratory why its dissolution profiles differ between two analysts and the answer is almost never the instrument. It is the medium. Degassing, temperature at the moment of filling, and the accuracy of the fill volume are all done by hand in most Bangladeshi plants, and all three shift the result.
Dissolved air is the worst of the three. Bubbles cling to the tablet surface and to the vessel wall, change the effective surface area and produce a profile that is reproducible for one analyst and not for another. Any laboratory arguing about method transfer should look here before it looks at the paddle.
The Media Preparation MediPrep Series (MediPrep 820, MediPrep 1622) exists to take that variability out. It mixes, heats, degasses and gravimetrically fills vessels in a single pass, handles foaming media such as SDS, and prepares up to eight litres in under fifteen minutes on the 820 or a larger volume in 25 minutes on the 1622, which Erweka put at three times faster than manual preparation to USP. The 1622 adds two inlets for premixed media, a waste water outlet and an automatic cleaning process that prevents cross contamination between products.
There is a second reason this matters in Bangladesh specifically. Purified water quality varies between plants and across the year, and a dissolution profile is sensitive to what is dissolved in the water before anything else is added. If the laboratory water system is not producing a consistent grade, the dissolution bench will report it as instrument drift. The grades and treatment trains are covered in choosing the grade, the treatment train and the consumables for laboratory water.
Dissolution automation is usually sold as a single leap from manual to robotic. In practice it comes in four steps, and most Bangladeshi plants should stop after the second.
Where the assay is by chromatography rather than absorbance, the Dissolution HPLC On-/Offline System is the equivalent path, automating up to five of the eight dissolution steps by pairing a DT 950 or DT 9510 with CTC sampling and online HPLC. Filtration at 0.45 micron runs through the automatic filter changer AFC 825 alongside the maintenance free PVP pump.
Be honest about which step your laboratory is at. Step four needs a service engineer who can reach the plant quickly, a stable supply and an analyst who understands the whole chain well enough to diagnose it. A plant in Gazipur running two shifts with a technician who has been trained on the instrument gets more from step two, reliably, than from step four intermittently.
Disintegration testing is mechanically simple and procedurally fiddly. A basket rack of six tubes is raised and lowered in heated medium, and somebody watches for the moment the last fragment passes the mesh. That last part is the problem: the endpoint is an observation, and observations vary.
The Disintegration Tester ZT 120 Light (ZT 121 light, ZT 122 light) is the entry point, with one or two motor driven stations, an integrated flow through heater and a one piece moulded PET bath. The run time counter starts automatically as the basket rack is lowered, which removes one manual step. The Disintegration Tester ZT 320 Series (ZT 321, ZT 322, ZT 323, ZT 324) drives one to four stations individually, each with its own keypad, lowers the rack automatically at the start and optionally raises it at the end so disintegration stops when timing stops.
The Disintegration Tester ZT 730 Series (ZT 731, ZT 732) removes the observation altogether. Its cable free AirBasket uses magnets and sensors to determine the disintegration time itself, with a sealed electronics compartment that makes the basket dishwasher proof and sensors that tolerate acidic media. Type A takes six test tubes for standard tablets and type B takes three for larger ones, both to USP, EP and JP. The seven inch touchscreen holds up to 200 products or methods with their parameters and results.
For a plant running many products and rotating analysts, the automatic endpoint is worth more than the extra stations. It converts a subjective reading into a recorded one, and the recorded one survives an audit.
Friability is the cheapest test on the bench and the one most often run slightly wrong. The pharmacopoeial method calls for the drum to sit at ten degrees, and a laboratory that has lost the stand and set the drum flat is producing numbers that cannot be compared with anyone else's. The Friability Tester TAR II includes that stand as standard, which is not a marketing point so much as a reminder that it matters.
Beyond the geometry, the TAR II is upgradeable in a way that suits a growing plant: one or two test stations plus an optional smart scale, all of which can be retrofitted later. TestAssist offers three modes covering revolutions, time and the USP test, checks that each parameter is set correctly, and in USP mode calculates the result and triggers a repeat where the guidelines require one. The balance connection transfers weights directly, which removes the transcription step where friability results are usually corrupted.
One practical note for Bangladesh. Friability results move with ambient humidity, because the tablet surface takes up moisture. A laboratory that runs its friability test beside an open window in July and beside an air conditioner in January will see a seasonal pattern that has nothing to do with the process. Log the room condition alongside the result and the pattern becomes explicable rather than alarming.
Breaking force testers divide by how much they measure and how much of the handling they do.
The honest advice is to buy the level of automation matched to sample count, not to ambition. A plant testing twenty tablets a shift does not recover the cost of a fully automatic combination tester, and it will lose the ability to test odd shaped products that a manual station handles without complaint. A plant running continuous compression across several products does, and the saving is analyst time rather than accuracy.
Half the causes of a bad tablet exist before the tablet does. Flow behaviour decides die filling, and die filling decides weight variation, which shows up later as hardness scatter.
The Granulate Flow Tester GT Series (GTL, GT, GTB) covers this in three levels: the GTL measures flowability by weight and by volume to USP and EP, the GT adds a scale so weight is followed during the flow along with flow time and flow angle, and the GTB adds automatic laser measurement of the angle of repose. The Tapped Density Tester SVM II works out the Hausner ratio and compressibility index automatically and prints a report after each test, with a SwitchPlate system giving five instrument variants covering USP 1, USP 2 or USP 3 in one or two station configurations, changeable within minutes. For a simple apparent bulk density figure to DIN ISO 697, the Manual Powder & Granule Density Tester SMG 697 does the job without electronics.
Suppositories are a separate small family that plants making them tend to under equip. The Suppository Penetration Tester PM 30 is Apparatus A to EP with the PM 3 as Apparatus B, the Suppository Disintegration Tester ST 35 gives three turnable stations in four litre vessels with a bath held to plus or minus 0.2 degrees, the Suppository Hardness Tester SBT 2 loads weights until the suppository collapses, and the Melting Point Tester SSP measures melting point in a spiral glass basket inside a water jacketed chamber.
Packaging integrity is worth a line as well. The Vacuum Leak Tester VDT/S runs the methylene blue dye test on blisters through a programmed vacuum, hold and penetration sequence with the vacuum documented over USB, and its integrated pump can be recalibrated with the optional validation kit. In a country where blisters travel long distances in heat and humidity before reaching a dispensary, a seal integrity test is not an optional refinement.
Every instrument above produces a number that will be looked at by somebody who was not in the room. How that number gets from the instrument to the report is what an inspection examines, and it is where most laboratories are weakest.
Three levels exist in the Erweka range. Dissolution Tester Software Disso.NET 4 controls every function of the connected devices, from offline systems up to the RoboDis II+, and documents what, who, when and why for every change in an audit trail that can be searched and filtered, with Active Directory login. Physical Tester Software MC.NET does the equivalent for hardness and combination testers, holding data from several devices in one networked SQL database with electronic signature in place of handwritten approval. Physical Tester Software Export Manager is the pragmatic middle option: installed on any PC on the laboratory network, it receives reports from devices such as the SVM II, TAR II, EasyCheck, MultiCheck 6, ZT 720/30 or TBH II and converts them automatically to PDF and XML, named with device type, serial number and the date and time received. Scanning printed reports stops.
Two practical points about installing this bench in Bangladesh. First, the water bath instruments in this range hold temperature by a combination of heater and thermal mass, and a bath that loses power for an hour has lost the run rather than paused it. Put dissolution and disintegration on protected supply and confirm the changeover works with the bath at temperature, not empty. Second, calibration is a recurring commitment rather than a commissioning event: vessel verticality, shaft wobble, rotation speed, bath temperature and balance performance all need a documented schedule, and the reasoning behind building one that survives scrutiny is set out in calibration and traceability for laboratory quality systems.
Get those two right and the rest of this equipment is undemanding. It is bench mounted, it draws little power, and with reasonable care it runs for a very long time.