We have some hign ampage rack switch mode power supplies - upwards and over 300 Amps. Through reserach the level of 2 joules have been considered high enough to cause burns. Just to mention there is no energy but electrical - and due to the very long runs we have turned up the voltage over 30 volts because of line loss.
I am wondering if LOTO for these racks should be used when working on down stream devices.

Given human bodies are around 1000 ohms at 30 VDC (resistance does drop as voltage increases) as per IEEE standard 80 and ohms law gives I=V/R so 30 / 1000 = 30 mA. Enough to be painful but not lethal shocks.

As to arc flash Ayrtons equation gives a lower minimum of about 28 VDC to get an arc at any available fault current. At 30 VDC you would need several thousand amps with any appreciable air gap such as 1/4". With no appreciable arc, arc flash is not possible. So no hazard there either. Don't get me wrong you do weld at these voltages but the trick is to make contact and then pull back 1 or 2 mm. Too close and it just sticks. Too far and the arc goes out. This is much different from anything that can exist uncontrolled.

Some tests have been done at 130 VDC concerning 125 VDC substation batteries by Duke at Kinetrics. At 1/4" gap at 20,000 A available fault current they achieved an arc for 80 milliseconds which is right at the 1.2 cal/cm2 threshold. So based on your description at both much lower voltage and current it seems pretty unlikely even on a fault in a capacitor. At 30 VDC you'd need total resistances under 1.5 milliohms...seems highly unlikely.

For those folks on the forum here North of the border, here is the not so exact wording from CSA Z462. I am sure that it is the same or similar in 70E

Equipment operating at less than or equal to 30 V
When energized electrical conductors and circuit parts are operated at less than or equal to 30 V they shall not be
required to be de-energized when the capacity of the source and any overcurrent protection between
the energy source and the worker are considered and it is determined that there will be no increased
exposure to electrical burns or to explosion due to electric arcs.

Again the reference to amperage. This is where, as Paul has referenced our good old friend Ohms Law. The average human resistance - dry intact skin is approximately 1000 ohms. If one happens to know the voltage then I (amperage) is equal to E (voltage) over R (resistance) if the calculated amperage could be dangerous then it would be best to Lockout