As David Barker pointed out, you need to remove primers and dNTPs from the mix before you sequence. The easiest way to do that is to treat it with Shrimp Alkaline Phosphatase (aka SAP - it degrades the dNTPs) and Exonuclease 1 (aka Exo1 - it degrades any single stranded DNA, including primers). There are commercial mixes of the two enzymes sold as kits (eg. ExoSAP-IT from Affymetrix) but you can also buy the enzymes separately (this is usually slightly cheaper, especially if you buy in volume). It's a simple protocol - the PCR reaction provides adequate buffer conditions for the enzymes so there's no other reagents needed. If you do a google search for SAP/EXO you'll find dozens, but here's one protocol that's similar to the one I've used in the past: http://www.protocolpedia.com/component/sobipro/?pid=69&sid=2535:PCR-SAPEXO-Cleanup-of-PCR-Products&Itemid=0
The other method we've used is to cut the band out of the gel and use one of the gel-degrading enzymes for cleanup (we use Gelase from Epicentre). Again, it's slightly cheaper and easier than column based methods, and provides adequate template for sequencing or TA-cloning.
The problem with unpurified PCR product is that both primers are still present in large amounts, so your sequencing reaction will include products from both strands and will likely be quite unreadable. So you need to purify the product away from these residual primers even if the product looks clean. There are many relatively inexpensive reaction clean-up kits such as:
You can use gel filtration columns as a homemade solution or possibly silica beads, but you will still need to buy all of the appropriate reagents and spend time optimizing these methods, so you won't necessarily save money and you definitely won't save time.
As David Barker pointed out, you need to remove primers and dNTPs from the mix before you sequence. The easiest way to do that is to treat it with Shrimp Alkaline Phosphatase (aka SAP - it degrades the dNTPs) and Exonuclease 1 (aka Exo1 - it degrades any single stranded DNA, including primers). There are commercial mixes of the two enzymes sold as kits (eg. ExoSAP-IT from Affymetrix) but you can also buy the enzymes separately (this is usually slightly cheaper, especially if you buy in volume). It's a simple protocol - the PCR reaction provides adequate buffer conditions for the enzymes so there's no other reagents needed. If you do a google search for SAP/EXO you'll find dozens, but here's one protocol that's similar to the one I've used in the past: http://www.protocolpedia.com/component/sobipro/?pid=69&sid=2535:PCR-SAPEXO-Cleanup-of-PCR-Products&Itemid=0
The other method we've used is to cut the band out of the gel and use one of the gel-degrading enzymes for cleanup (we use Gelase from Epicentre). Again, it's slightly cheaper and easier than column based methods, and provides adequate template for sequencing or TA-cloning.
I agree with both of my colleagues. Contaminating dNTPs will result in top heavy sequence and contaminating primers will cause a mixed and either noisy trace or at worst a garbled ladder.
Maldonado like them I have cleaned up with column based methods but found SAP Exo digestion to be a cheaper and effective alternative