• Question: how many times has an experiment failed,what was the experiment

    Asked by thomascockram9 to Lyn, Katy, Paul, PB, Ruth on 21 Jun 2013.
    • Photo: Peter Balfe

      Peter Balfe answered on 21 Jun 2013:

      Anyone who can number the times an experiment has failed hasn’t been in the lab very long!
      My “best” failure was a ultracentrifuge head which came apart at 140,000 rpm one night. The armour plated bowl held the explosion, as it was designed to. But it was chilling to come in the next morning and see a 750 Kg machine on it’s side!

    • Photo: Ee Lyn Lim

      Ee Lyn Lim answered on 21 Jun 2013:

      I’ve just had to repeat an experiment for the 4th time today, and I still haven’t been able to get results (I’ll start planning for version 5 tomorrow…). The experiment should go like this: I take a mouse and give it a vaccine, so that its immune system becomes angry and active and ready to kill. 8 days later, I inject the mouse with some cancer cells, and see if the immune system manages to get rid of them.

      First time around: I couldn’t find the cancer cells again! Where did they go?
      Second time around: I tried just injecting the cancer cells without the vaccine, to see if I could find them later – and I found them alright, but there was no vaccine, so no results.
      Third time: I injected both the vaccine and the cells, but this time I got ALL the cells back – the immune system hadn’t killed any of them at all. Fail!
      And today: I decided to take the immune cells out of the mouse and treat them with vaccine directly, then inject them back into the mouse together with the cancer cells and see if that would work. But guess what – the injection failed. The cells kept coming back out again! I can’t for the life of me figure out why…

      There are a hundred different things that can go wrong in any experiment…and most of them happen at some point or other!

    • Photo: Ruth Mitchell

      Ruth Mitchell answered on 23 Jun 2013:

      Many experiments have failed!! I have just done a third repeat of injecting some mice which is a 3 week protocol and it has given me completely different results to the last 2 – so going to have to do that again!

      Experiments can “fail” either because of biological variation(mice being different like what happened with me) or because of human error(me being silly!). But sometimes it comes up with interesting results like finding you can use a drug at 10 less concentration and get the same results!

      Someone in my lab said: experiments can go “into the void of the unexplained” – it’s on a post-it note on my computer!

    • Photo: Paul Waines

      Paul Waines answered on 25 Jun 2013:

      Lots- it happens to everyone. When it happens you have no choice but to try and understand why, and get on with it again.

      Sometimes it is because of a silly mistake- for example when I blew my sample up in a microwave (and the microwave). Other times, it is for no apparent reason, and these are the most frustrating. However, going through this is an important part of learning and ganing experience.

    • Photo: Katy Brown

      Katy Brown answered on 25 Jun 2013:

      Yep, experiments fail all the time, probably more often than they work! I left a very small typo in a program and used the results it generated for the next two years, turns out they were all wrong. It’s very frustrating but you just have to try again…

Comments