as a general advice you need to be gently with platelets because mechanical forces (rough pipetting, shaking…) may activate them. I recommend to turn off the break of the centrifuge (braking level to 0 or at least use the setting “soft”).
For detailed information I need to know which species the blood is from. Which anticoagulant are you using (heparin, sodium citrate...)? Do you want to use platelet-rich plasma (PRP) or washed platelets?
centrifugation is really important in platelet preparation. We used a first round at 150 g to pellet erythrocytes and leukocytes while maintaining platelets in solution = platelt-rich plasma, followed by a second round at 300 g to pellet the platelets. Each time without brakes on.
When really every stimulation of the platelets should be avoided, I would recommend whole blood flow cytometry. In FSC-SSC plots the platelets will be right in the middle between leukocytes and erythrocytes. If you want to be totally sure to analyze the right population just add a platelet-specific antibody for identification, e.g. CD41 and/or CD61 (only present on platelets and megakaryocytes) and separate the right cells by this one. Then analyze your surface receptors of interest... this worked very well for us and I hope will do also for you!
You need to be really careful with platelets! You can find a really robust assay for platelet activation in our paper "Platelet activation in essential hypertension during exercise: pre- and post-treatment changes with an angiotensin II receptor blocker."
Anticoagulants and the way you treat your samples all depend on what you want to study with platelets. If you are going to study surface markers for identification of platelets (Diagnosis of Bernard-Soulier Syndrome or Glanzmann, or even pAIg) you can use EDTA blood, separate platelet rich plasma (PRP) and stain them. But if you are going to analyse activated platelets, special care is needed, you need to collect into heparin or sodium citrate tubes, blood should be drawn without using a tourniquet, first 0.5 - 1.0 mL should not be used, a second tube should be used, and a whole blood staining protocol should be used.
I would recomend using a platelet marker such as CD41 which is the most abundant receptor on platelets, and also as metioned before specific to platelets. Further I would recomend using staining for P-selectin (CD62P) in your samples (or in a separate control sample) to get an idea of how activated your platelets have become from just handling them. This is good for whatever type of platelets you are using (whole blood, PRP or washed platelets) but especially good when you're learning how to wash them as it will become obvious if you have been gentle enough or not. Good Luck!
Further the need for buffers and such depends on how you want to separate your platelets. For PRP, you do not need anything, just spin them and discard the red cells. For washed platelets it depends on what you want to do with the cells. ACD is normally used to avoid activation in the centrifugation step, further you can add inhibitors like apyrase or PGI2 as well, but again what is appropriate depends entirely on what you want to look at in your platelets. Obviously, the same goes for the anticoagulant used.
As everyone else has said platelets are quite sensitive and react easily to manipulation. So if possible it is best to reduce handling them as much as you can. The link below is to a consensus protocol for platelet flow cytometry, it's a bit old now but still very valid.
http://www.ncbi.nlm.nih.gov/pubmed/9609215
If you want to look at basal levels another anti-coagulant that you can use is CTAD; this can't be used for activation studies but is helpful if you want to prevent any activation occurring. EDTA is not typically used for platelet activation studies since the EDTA can affect the glycoprotein structure on the platelet surface, the most common is citrate or variations of this like ACD for washed platelets.