capacitor expert needed

I am working with a sensitive prototype electrical module that fails under voltage fluctuation (voltage drop specifically). I need to make a circuit inside a small project box that will solve this problem, because the module cannot be changed and must be used. The module draws as much as 15 amps. It seems clear to me that capacitors will buffer voltage drops caused by sudden current draws in the vehicle. I need to determine what type of circuit will give a high enough capacitance to accommodate the current requirements while providing the fastest discharge rate possible. One large cap, many small, etc... (isolation is not an option) I'm always impressed at how much more people on this forum know than engineers when it comes to actual electronic components. Thanks for the help guys.

hi, are your power supply and ground wires of adequate size? how much of a voltage drop is present? is the voltage drop measureable in the module itself? if the unit is powered with more than adequately sized wiring, direct from the battery, does it have the same failure? does the failure occur whether the vehicle is running or not? does the unit fail on the test bench? mark

"Ground wires of adequate size": Yes, its a production vehicle, and all wires are tapped into vehicle wiring. Voltage drop: very small, less than a volt (equipment is very sensitive). Yes, the module is connected to the voltage source, therefore it is susceptible to all variations. (It is wired directly to the battery, Power and Ground). (Also, don't suggest different grounding, been tried extensively) Test bench is not an option because this is a remote problem (bravo for the suggestion though, it's what I've been pushing for), (not an option though). hi, are your power supply and ground wires of adequate size? how much of a voltage drop is present? is the voltage drop measureable in the module itself? if the unit is powered with more than adequately sized wiring, direct from the battery, does it have the same failure? does the failure occur whether the vehicle is running or not? does the unit fail on the test bench? mark

the module has a control or input section, and an output section? are the power feeds of the sections independent? a diode/cap or resistor/cap on the input section power feed should decouple it from the output.

Onya Ween. Frank - there is something wrong with that prototype design. Apart from NOT tapping into the vehicle's wiring but being taken directly from the battery (both + & - terminals), it should be able to operate with more than a 1V drop. I usually design stuff to operate from 8V to 16V (else tolerate such "typical" voltage fluctuations). What are the voltage parameters for this prototype? Alas not knowing more, I cannot advise. It may simply require in/out decoupling as Ween suggested. Or it may require a redesign or a dc-dc front end... Or its own battery. I mention the above in case it is a production prototype (before it loses $millions!). Also in case it is a one-off bad design. (I have seen so many. And accurately predicted so many failures, And bankruptcies.)    Can you describe its purpose or function? Or components (digital, analog, motors, sensors?). (I presume sensor grounds are separate from digital and power grounds, and analog etc - ie a single point of interconnection else differential inputs?) And IMO a 1V drop is too high, but I have said before that I usually try for under 0.5V TOTAL drop (+ & - paths) but some seem to feel that 3V is acceptable. (Even a the12volt.com reference source recommended a ground resistances under 1/2 Ohm - that's a 7.5V drop at 15A! I'm sure it meant to say 0.5mOhms!)

Without knowing more info it is hard to offer many suggestions. Basically, though, it sounds like you need a fully regulated power supply. At 15A, however, it will be difficult to build a fully regulator power supply AND store it in a small box! What in the circuit draws 15A? Is it possible to run the main/input part of the module at 9vdc and use relays to control the 15A load? Kevin Pierson

The vehicle is in another State and there is not a technician on site to do all the testing and reground that should be done. The decision has been made to develop a circuit as described above, I just have to build it. Its clearly the wrong fix, but its what the customer wants. Does anyone have any advice on the best way to approach this "bandaid?"

There is nothing to approach - you have to supply more information. Otherwise get a BIG cap or another battery to "filter" the supply. What else can you do?

Nothing, I'm trying to find out if using one large cap would work better than using multiple small caps. Which would have a faster charge/discharge rate to absorb low-medium voltage drops?

They are equivalent. The only difference is in their technology and maybe distribution - eg tantalum vs electrolytic; and one big big versus distributed small. And of course a (small & cheap) AGM maybe better, but since the customer is paying....