TSS Troubleshooting & Quality Control

TSS is generally a fairly simple test with few trouble spots. The most common mistakes involve filter preparation and sample mixing. Running duplicates on a continuous basis provides helpful information and builds confidence in the technician's procedure.

Blanks are inconsistent
Methods blanks should be performed on a routine basis, a least once per batch of filters prepared. The method blank is the same as re-rinsing the filterswith 100 ml of deionized water. If the filters have been washed properly and consistently, the filters should not lose particles and should have an initial and final dry weight difference close to zero. If there is a difference of >0.0002 g or duplicate filters are not consistent, the filters may not have been rinsed completely.

For instance, if two method blanks are run and the first drops by 0.0001 g and the second drops by 0.0012 g, the difference may indicate something is wrong with the way the filters were rinsed. Method blanks can be run with each set of samples to show that rinsing is effective and that sample is not being carried over.

Not Enough Sample
The selection of the "right" sample volume is often a guess. The sample volume should ideally add a significant amount of solids to the filter to minimize any balance errors.

That is why the minimum weight is supposed to be less than 0.0025 g. Analytical balances have an electronic error of  +0.0001 g. This uncontrollable instrumental error would be a 10% error if the amount of solids was 0.0010 g. The error caused by the electronic noise goes down as more sample is added to the filter. When at least 0.0025 g of solid have been added, the error is now a reasonable 4% and will get better as more weight is added.

Always add as much sample as possible. If the sample is still filtering without clogging, and the filter looks clean, add more well mixed sample.

Figure: This filter may not have enough weight. Increase the sample volume to reduce error.


Too Much Sample
This is a problem just the opposite of the last one. In this case, too much sample has clogged the filter and the remaining sample is just slowly dripping. Several problems result. First, the filtration time becomes excessive. If the sample is still filtering after 10 minutes, discard the filter and repeat using a smaller volume. Second, a clogged filter will interfere with the rinsing step. The rinsing step should recover any straggler solids from the sides of the graduated cylinder and funnel and rinse out any trapped TDS. Trapped TDS will give a positive bias to the answer. Third, the excessive solids trapped on the filter will form a hard crust on the filter and may prevent water from evaporating completely.

This is visible when the filter has a lot of crazing or cracking on it. This is similar to clay. Clay looks dry but when broken into pieces, the inside is still wet. This water weight will also give the sample a positive bias.

Figure: This cracking indicates too much sample was filtered. Reduce the volume next time.

Figure: This sample was poured slowly and results in poor distribution of the solids. High solids in one area may not dry completely.


Duplicates are inconsistent
TSS samples should always be run in duplicate. It takes very little time to run a second sample. If the duplicates are consistent, the technician's technique,sample handling, mixing, rinsing, etc. is validated. If the duplicates are inconsistent, sample handling problems may be present.

Each sample measured should be mixed and poured completely. For instance, if 67 ml of sample has been measured, all 67 ml should be passed through the filter and rinsed. The sample should not be split, say 25 and 42 ml. While the average of the sample may be correct, the duplicate answers may be far apart. The 25 ml sample may have a TSS answer of 50 mg/L while the 42 ml sample may have a TSS of 84 mg/L. This large difference indicates the solids in the sample settled between the 25 and 42 ml pouring. It would be better to run a 30 ml sample all at once then mix and pour a second sample of 35 ml. The answers for these two separate samples will be more consistent. Homogenizing the sample may also help improve the precision.


ODD Stuff
In spite of the best effort to mix the sample, some samples contain odd stuff such as bugs, twigs, grease balls, etc. These odd particles can give very positively biased samples (high answers). Standard Methods allows you the technician to decide if these materials of truly representative of the sample. If the sample contains a lot of swimming critters, then perhaps they should be included in the TSS filter. If there is only 1 or 2 mosquitos and the technician pours one out onto the filter, perhaps this mosquito should be removed. The same argument applies to grease and oil. Grease and oil stick together and cling to the top of the sample bottle. It is often difficult to adequately mix these materials, so they get pipetted or poured off into the graduated cylinder because they are normally floating at the top of the sample. Is a large grease chunk on one filter representative? Is it a grease chunk? The technician may wish to run the sample and make a note of the abnormal particle.

Note: Large particles determined by the lab technician to be "unrepresentative" must be removed prior to analysis. They cannot be removed fromthe filter after filtration.



Quality Control

Decisions on the accuracy of the reported data will be based on the quality control information.


Sample QC

1. Sample holding time cannot exceed 7 days.
    Corrective Action: Reject samples and request a resample.

2. Samples must be preserved on ice or refrigeration until time of analysis. Record the
    temperature of the refrigerator.
    Corrective Action: Adjust refrigerator to below 6 oC. Service the refrigerator if the
    temperature does not adjust properly

3. Samples must be warmed to room temperature prior to TSS analysis.

4. Run samples in duplicate 100% of the time if possible. Use approximately the same volume
    of sample for both. Remix sample between tests.

5. Record sample date, time, type, sampler, date and time of analysis, analyst and method
    used.

6. Samples with large chunks of nonhomogeneous materials should be homogenized for 1-2
    minutes for better precision and accuracy. Avoid excessive homogenization which might
    cause volatilization of some solids.

7. Samples must be mixed well and poured quickly.


Equipment QC

1. Drying Oven must be 104 oC +1.0 oC

2. Record the temperature of the drying oven.
    Corrective Action: Incubator outside the control limits must be adjusted. An oven
    temperature below 103 oC may not dry the sample completely. An oven temperature above
    105 oC may cause some organics to volatilize.

3. Immerse the oven thermometer in sand to prevent inaccurate temperature readings when
    the oven door is opened frequently.

4. Calibrate the oven thermometer at least annually against a NIST certified thermometer.
    The calibration must include date, thermometer correction factor, serial number, and
    initials of the person performing the calibration.

5. Record the calibration data

6. Use an analytical balance capable of weighing 0.0001 g.

7. Calibrate the analytical balance annually using a certified balance technician. Document
    date, balance condition, and name of technician and company.

8. Calibrate the analytical balance at least monthly using Class 1 weights. Select a series of
    weights which covers the range of balance operation. Usually 1, 2, 5, 20, 50, and 150 g
    weights are used. Record weight values on Balance Log Corrective Action: If the weights
    deviate more than 0.0002 grams, the balance needs service. Use another calibrated balance
    until the next service cycle if possible



TSS Test QC

1. Pre-wash filters with deionized water and perform at least 2 method blanks on each lot
    washed.
    Corrective Action: If the weight difference is >0.0002 g, the filter has not been washed
    completely. Allthe filters in this lot must be rewashed and the process repeated until the
    difference is within 0.0002 g Document all method blanks.

2. Store pre-washed filters in the desiccator to avoid water absorption.

3. Zero the analytical balance prior to each weighing series Document the balance was zeroed.

4. Use large bore pipets for small sample volumes and graduated cylinders for large volumes.

5. Pipets and graduated cylinders are rinsed with deionized water and the rinse is added to the
    filter after the sample has been filtered.

6. Use sufficient sample to obtain a minimum of 0.0025 g of TSS on the filter.

7. Filters are checked for dryness. The difference between the 1st and 2nd dry weight is less
    than 0.0005g

8. Filters are placed in the desiccator upon removal from the drying oven. Document the
    condition of desiccator.

9. Performance evaluation samples should be run at least annually.

10. Split samples can be run with other nearby facilities

11. Duplicate sample TSS results should be within 10% of their average.
    Corrective Action: Homogenize samples with large amounts of chunky suspended solids to
    obtain a more uniform sample. Remix samples between duplicates and measure quickly.