The Bloom Box: a power plant for the home

bhankiii,

I don't know if you have ever done government work, but it's easy to see how this stuff happens:

The government has a requirement for an implement to drive a nail into a board. The contractor's engineer specs out a $3 hammer from a Home Depot supplier, and prepares the drawing. Adding profit and documentation charges, the hammer would cost $6 to the government.

The government QA guy, Mr. Covermyass, insists that first article testing is required before he signs off. First article testing is done at a cost of $60,000, and the hammer is approved. The government orders 10,000 hammers to be delivered over 2 years. So now the hammer costs $12 to the government, as part of a cost plus project.

In the second year, the hammer manufacturer improves the design by using a better metal alloy in the head. The contractor's incoming inspection flags the first shipment as not meeting the print. Since the change is an improvement, the contractor's engineer prepares a waiver/deviation document: "Does not effect form, fit, or function. Recommend use as is". The waiver is submitted to Mr. Covermyass.

Mr. Covermyass comes out of his office yelling that the contractor is trying to pull a fast one. "This is an unauthorized design change, not a deviation!" He refuses to sign the waiver. He insists that the drawing must be changed and first article testing be repeated to accept the hammers.

Since there is not enough time in the contract to do this, the contractor must go to specialty manufacturer to have the hammer made to the print. Because of set-up costs for the short run of this now "custom" hammer, the specialty manufacturer charges $700/hammer. Since it is a cost plus project, the government absorbs the higher cost.

Mr. Covermyass goes to sleep at night thinking how lucky the American public is to have such a good civil servant as himself to protect them from the evil contractor.

The next morning, Mr. Covermyass finds a drawing for a finishing nail on his desk for approval. He notices that the alloy is not called out. He mutters under his breath "Those bastards, they're trying to pull another fast one!"

Sorry for the rant. Too much coffee today.

Tad

My experience - 30 years worth - is not quite so bleak. What I typically see are situations were an off the shelf item just doesn't fit the bill. Maybe it's a wrench with an odd shaped handle to fit into an odd shaped bit of avionics. Maybe there's a requirement that the alloy not be brittle at temperature extremes, whatever. So the solution is to design a custom tool to do the job.

I've also seen cases where "Does not effect form, fit, or function. Recommend use as is" led to failure, since the guy deciding what the form fit and function was didn't realize the full intent of the guy who originally specified it.

I saw both sides as well, but I did see some crazy stuff.

We had a product that used ethylene glycol (EG), strictly as an antifreeze. It was mixed with 50% water before going in the product.

When the product was developed, the development engineer included on the print all the chemical specifications he found on the EG reagent bottle , including the specification that it contained <0.1 % water. Not a relevant spec, given the application. But, it was on the print.

As chemical suppliers became more sophisticated, they realized that there was no way to meet the water spec, as EG absorbs water from the air. So we couldn't find a vendor that would supply the EG with a Certificate of Compliance that it met the print.

My boss called in the government inspection QA guy, explained the situation, and asked to change the print. The QA guy was reasonable. He said he would not authorize a print change because it would would have to go through several approvals on the government side, which was always an expensive hassle, and that someone in the government chain might want cover their ass, and require a new first article test on the whole product (hugely expensive). He realized the spec was silly, and told us he did not care, just to use the EG without a C of C. He wasn't going to open a can of worms for such a stupid issue. As long as the product met the final functional test, he did not care about the cert.

My boss's position was that we can't violate QA rules on the inspector's word. If audited we could be found to be out of compliance, and open to recall of $200k of product or more, debarment etc.

Both sides were right, and would not budge. I was stuck in the middle.

My solution? We ordered the EG without the C of C. When it came in, our internal QA would give me a call, and I would go up and pull a sample. I took it to my lab, and poured it down the sink. An hour later, I would submit an analysis report certifying that it contained <.1 % water. It was then accepted by QA.

One day I'll burn in government hell, but I feel I did the right thing.