Hi All,
I am taking over a small wastewater treatment plant and have been reviewing historical data.  The plant must sample for BOD, TSS, DO and pH twice per month. The influent is tested for BOD while the effluent is tested as CBOD.

Once a month the laboratory splits a sample and sends it to an outside certified laboratory to compare results. When comparing the results, the % difference of the TSS and CBOD are minimal. However, the BOD results are not even comparable. E.G, while our lab may get a BOD result of 75 mg/l the outside laboratory will get 150-180 mg/l. The BOD results are consistantly 2-3 times higher from the outside lab.

Our plant laboratory does 2 dilutions on the influent BOD samples. The % difference in the results of each dilution is within acceptable ranges.  The two sample volumes we use for the influent BOD are 9 ml (diluted to 300mL) and 18 mL (diluted to 300 mL).

I am hoping someone can give me some ideas as to what we need to do to correct this. I am thinking that we may need to change the sample volumes used for the BOD samples.
Thanks for any ideas!
Chris

Chris,
That is indeed an interesting predicament.  If the outside lab had been lower,  I would have suspected them of accidentally doing a CBOD.  A little bit more information would be helpful to evaluate what is going on.  What are your blank depletion values, G:Ga values and seed control values?  What were the initial and final DO of the samples that you did?  See if you can get the same information from the other lab on what they reported to you.  There is a possibility of a toxicity effect, but you would have to do 7-9 different dilutions to really see that.  You would need to choose the dilutions to be somewhat close to one another.  I would range maybe from 0.5 mL to 30 or 50 mL.  That would give you an idea of if the sample itself was having an effect on the test.

The difference in results may be how the samples are treated prior to the BOD determination. The results from your lab sound like filtered samples and the certified lab's samples sound like normal influent. If the certified lab is nearby; then have the two analysts get together and observe each other's analyses (I realize that each analyst is absolutely sure of their results, but some details may be overlooked). If this doesn't work, Then try adding a known quantity of BOD (with some organic salt) to each of the samples to determine the recovery of each additional BOD.

grrun

Thanks to everyone's response. I am waiting on a call from the outside laboratory to find out information on their dilutions/preparation steps. I will post when I hear back.
Thanks
Chris

My first observation is that you don't have enough initial DO in your blank. The method specifically requires that you have a minimum of 7.5 mg/L and maximum of 9 mg/L DO present in the dilution water.  This obviously won't affect your samples, but I wanted to point it out.  The lab I worked in also had the same requirement for samples.  This isn't in the method, but it gives you more room to work with.  Usually some vigorous shaking of the sample in a half full container will bring the DO closer to saturation.  Secondly I notice that you are not using (or at least did not report) any kind of seed correction.  You may not have a healthy enough population of bacteria in your samples to give a proper BOD.  Try using seed bacteria of some sort the next time you do your BOD.  That should improve the performance of the test.  You could try various spots in your treatment plant and see how they work or just buy some commercial freeze dried seed.  Whichever you prefer.

Yeah, you need to get more oxygen into that initial DO reading.  This test must be run with stringent QA/QC limits to get consistent and valid results.  It is a pain, but that is just the nature of the beast.  

This does not appear to be a case of toxicity.  If it were then the other lab would be dealing with it as well.  Set up procedures would appear to be at the root of the problem.  Meter calibration could be causing faulty DO readings.  Or your dilution water may be "off."  But then, you say the final effluent results come out alright.  I still say doublecheck how much mixing of the sample gets done before those two dilution volumes (9 and 18 mls) get drawn off.  Your lab is not doing something that the outside lab is doing.  You simply must identify that "something."

~Eric  Virginia is on the upslope of the weather "rollercoaster" for the next couple of days.  Break out the shades and light jackets!

Just an update on this issue. After much cleaning/sterilizing, checking QC and overall laboratory procedures I have gained some ground. Last month when I ran my second set of samples, I also split the set with an outside certified laboratory.  After comparing the results, here are the percent differences in our numbers.
Influent
BOD = 5.6 %D
pH = 2 %D
TSS = 8 %D

Effluent
CBOD = 2.1 %D
pH = 0.5 %D
TSS - Not applicable. The laboratory reported <4.0 and I had 3.3.

Overall I am happy with these numbers and appreciate everyone's input.

Thanks
Chris