When I switch to SQUARE wave its where most of distortion is. I think my connections were bad when I first posted.
What you are seeing on the square wave is called "ringing" and it is caused by the way you have tried to probe the signal, not by the signal source or the bandwidth of the probes you are using. I'm guessing that you used the 3 inch or 6 inch ground leads that came with your probe to connect to the signal ground. That lead is adding a lot of inductance to your measurement and causing the ringing you are seeing on after the rise and fall of the square wave. If you really want to reduce the ringing you need to reduce the length of your ground loop, either by using a ground spring (which probably came with your probe) or by using a probe-tip to BNC adapter (if you are getting your signal from a BNC connector).
Here is a good video that spends some time talking about probe grounding techniques:
As to the question about using high bandwidth probes on a low bandwidth scope: it's better to use the probes that came with your scope, but if the "high bandwidth" probes are of the cheap Chinese variety, it may not be making all that much difference. Still, it would be better to get probes that are properly matched to the scope, both in bandwidth rating, and in capacitance.
Probes have a compensation network that allows you to match the probe to the instrument. Most of the compensation adjustment is done by an adjustable capacitor, and the probe documentation should tell you what the range of that adjustment is. What you need is a probe whose compensation range contains the input capacitance of your scope. Your scope's documentation should tell you what the input capacitance is. As an example, I have an old Tektronix 475, a 200 Mhz analog oscilloscope, with an input capacitance of 20 pF. I bought cheap Chinese probes on Amazon rated at 200 MHz with an input capacitance adjustable from 13 pF to 25 pF, so those probes can be compensated to match my scope.
Higher bandwidth probes often have lower compensation ranges than lower bandwidth probes, and higher bandwidth scopes have lower input capacitance than lower bandwidth scopes. If the compensation range for you probe does not encompass the input capacitance of your scope you will not be able to properly compensate the probe, and you will get distortion of your measurements. So, best to get probes that are well matched to your scope, both in bandwidth, and in compensation range.
As for using your probes to filter out high bandwidth components of your measured signals, that is probably not an issue. While I am quite happy with the 200 MHz cheap Chinese probes I bought (you can't go far wrong for $15), other people have reported that the Chinese probes are actually much lower bandwidth than their specs indicate. So, even though you bought 200 MHz probes, you may actually have only 100 MHz probes. In order to test that you would need a much better signal generator than you probably have on hand, and since your scope is only 70 MHz, it's not even worth worrying about. Suffice to say that your 200 MHz probes are probably not causing your noticeable problems, assuming that you were able to properly compensate them (which you said you did).
-- Jeff Dutky