Yes credit frictions do have an amplifying effect on supply shocks while they often attenuate demand shocks and they also give increased persistent to the effects of the supply shocks in many models.
Once you have a specified DSGE model you can choose exactly where the shock originates from and the strength of the shock. See for instance Gerali et. al. 2010. Credit and Banking in a DSGE model of the Euro Area. They incorporate a banking sector into a traditional DSGE model. Their variance decomposition shows that it is mostly shocks originating in the banking sector which explain the fall in GDP during the financial crisis. However that is empirical estimation.
What you can do and what they do is to shock bank capital so as to simulate a loss. They show that this loss in bank capital has effects on output, consumption, inflation, investment and so on. So by incorporating financial and/or credit frictions you can simulate shocks to these which in turn should have real effects in your model.
The paper can be found here: http://www.bancaditalia.it/pubblicazioni/econo/temidi/td10/td740_10/td_740_10/en_tema_740.pdf
EDIT:
There are actually papers that are examining the question at hand in even more details than the Gerali et. al paper mentioned above. Two of these are Iacoviello, M. 2014. "Financial Business Cycles" where he examines a recession whcih is initiated by losses suffered from banks. Another paper is Jensen et. al. 2013. "Changing Credit Limits Changing Business Cycles" where they examine financial liberalization in a DSGE model. They find that financial liberalization leads to higher macroeconomic volatility.