PWM improves fuel cell efficiency? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

PWM improves fuel cell efficiency?

Move Thread

LAN_403

GammaRay

Sun Jan 13 2013, 06:46PM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

I have a PEM fuel cell that produces hydrogen/oxygen from water when v/amp is applied. I am curious if PWM of the power supply to the fuel cell would increase overall efficiency, or not. I have spent ample time surfing the net and other forums for a convincing answer. There is the camp that claims efficiency is increased (mainly due to improved thermal factors when using PWM) and camps that disagree stating overall efficiency remains the same with or without using PWM. Still looking for a solid answer.

What says this forum?

Pinky's Brain

Sun Jan 13 2013, 06:54PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

What's the point in using a fuel cell for electrolysis? It seems needlessly complex and expensive.

Carbon_Rod

Mon Jan 14 2013, 07:21AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Surface chemistry usually adds a dwell time for a catalyst to initialize an efficient reaction, and added heat energy will often improve the threshold for a reaction occurring.

PWM sometimes helps degas/clean reaction surfaces, but in general would not significantly improve power output*.

*Disclaimer: I said in general, as we assumed your chosen PWM frequency does not directly interfere with the surface catalytic reaction, increase cell internal resistance, or exceeding the sampling rate of instrumentation.

Thermodynamics general rules according to C. P. Snow:
0. "You must play the game."
1. "You can't win."
2. "You can't break even."
3. "You can't quit the game."

Ash Small

Mon Jan 14 2013, 01:58PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

From what I've read, the temperature inside a fuel cell must remain within certain boundaries in order to prevent damage.

'Some' wet chemistry benefits from increased voltages (increased power).

If PWM enables both of the above, it 'may' be advantageous in some circumstances (One example I have experience of where this is the case is electro-polishing of stainless steel, using 'certain' electrolytes), but I've no experience with fuel cells, so don't quote me.

Patrick

Mon Jan 14 2013, 05:23PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Carbon_Rod wrote ...

Thermodynamics general rules according to C. P. Snow:
0. "You must play the game."
1. "You can't win."
2. "You can't break even."
3. "You can't quit the game."

does everyone see how cruel and inhumane the universe really is?

2Spoons

Mon Jan 14 2013, 11:31PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Pinky's Brain wrote ...

What's the point in using a fuel cell for electrolysis? It seems needlessly complex and expensive.

It is useful having an electrolyte that is essentially solid. If the cell spills, it only spills clean, safe, water. I seem to recall that higher efficiencies are possible, due to the (usually) platinum catalyst. The solid electrolyte membrane also makes a really good gas separator.

As for the pwm, well things like diffusion rates, and saturation of the catalyst may come into play - so giving the cell a periodic rest may improve the cell impedance.
Only one way to know for sure - try it.

Steve Conner

Wed Jan 16 2013, 12:09PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

In electrochemical processes, the useful work done depends on the average current. However, the energy lost as heat depends on the RMS current.

It follows that all electrochemical processes are most efficient with smooth DC (which has a RMS-to-average ratio of 1) and degrade as the ripple increases. PWM is an extreme example of ripple, so I would expect the performance to be poor.

The same consideration applies to Peltier modules, indeed the ones I've worked with have a maximum ripple current spec in the datasheet. If you want to drive a Peltier module with PWM, you have to use a high frequency and add an inductor to filter out the ripple.

Ash Small

Wed Jan 16 2013, 03:27PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Steve Conner wrote ...

In electrochemical processes, the useful work done depends on the average current. However, the energy lost as heat depends on the RMS current.

It follows that all electrochemical processes are most efficient with smooth DC (which has a RMS-to-average ratio of 1) and degrade as the ripple increases. PWM is an extreme example of ripple, so I would expect the performance to be poor.

As a generalisation, yes, however with some processes, eg electro-polishing, for example, you need a certain voltage before electro-polishing occurs. below this voltage you get electro-etching instead.

As you turn up the current, you first get electro-etching occurring, and the voltage rises with current. then you reach a point where the voltage increases, but the current remains constant. you then reach a point where the current then starts to rise again with the voltage, and at this point electro-polishing starts. As you need to keep the temperature within certain limits (max temp of the 'recipe' I use is 205 C, above this the electrolyte isomerizes-at least I think that's the correct term), if you don't use cooling coils, you have to turn the power on and off to 'keep it cool'. Turning it on and off to maintain the temperature within limits amounts to PWM, so in certain circumstances, with certain electro-chemical processes, PWM is not only advantageous, but necessary.

I can't give precise voltages and currents, because each different sized and shaped item requires different power levels, as do different 'concentrations' of electrolyte.

It is, however, quite strange to watch the current rise with the voltage linearly, then there to be a 'step' in the graph, then the current starts rising linearly with the current again.

I have a theory as to why this is the case, but won't go into it here, at this time.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob