Peltier cells

[Aly]

Guys does any one have experience with working with peltier cells ?

[Aly]

Guys does any one have experience with working with peltier cells ?

[Warpspeed]

What are you trying to do ?

[BeunMan]

Probably trying to cool something down or heat something up using a lot of power.

You can also do the reverse.

[TMichaels]

I would not bother using them.
They are rather inefficient...

[Aly]

The case is that I've calculated area of radiator for our engine GSR600 and came out to be 1200cm2 .

So I decided we should use the radiator we have 1214 cm2.

The case is we have a turbo charger that cools by oil and has a water cooling backup system in case it over heats for any reason .
Plus we cannot rely totally on our old radiator because some team members reported overheating of engine sometimes .

So i thought of installing these cells in the water tank ... and make them cool water in case of over heating in engine or turbo (in case cooling water temperature rises above normal for any reason).

So it won't work all the time .... only they start working to save the day when overheating happens ..

I want to hear your opinions

Regards

[TMichaels]

I know that it sounds sassy, but are you aware of the principle of energy conservation?
Peltier elements become cold on one side and hot on the other side. Therefore you have to cool the hot side to achieve decent cooling on the cold side.

The main problem of most of the teams is air ducting and not the size of the radiator.

If you really have a problem with cooling your oil, I would think about cooling the oil separately with an extra radiator and without heat transfer between oil and water. The oil radiator will be more efficient due to the higher temperature difference and thus the total area of a water radiator and an oil radiator may be smaller than a single water radiator which needs to reject the energy dissipated in the oil and the water.

Please anybody correct me, if I am totally off. I am a sparky and I am only repeating what I picked up during my time in the team.

[Gruntguru]

Has any of your team mates enrolled in a thermodynamics unit yet?

If you want to cool something that is already hot (water or oil) you cool it by transferring the heat to the cooling medium available to you (the air). A heat pump (a Peltier for example) will pump your waste heat up to an even higher temperature (which you don't need to do) perhaps improving the transfer of the heat to your cooling medium. However the heat pump requires energy to operate and this must also be dissipated to the air - ouch! Furthermore, that electrical energy must be produced by the engine (at an efficiency of 20-30%) so the cooling water/oil heat load also increases (ouch again). Are you starting to get the picture?

[Warpspeed]

Agree with all of the above, Peltier cells are a very inefficient way to cool anything down to ambient.

The electrical power required would be far better used to power a monster electric radiator cooling fan along with some more efficient air ducting.

Until you actually track test the car, it is not really possible to anticipate what final water and oil temperatures will be.

Get it up and running, and do some logging of temperatures around the system. There may very well be no problem, or if there is a problem, the solution should be fairly self evident.

[Zac]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Aly:
The case is that I've calculated area of radiator for our engine GSR600 and came out to be 1200cm2 .

So I decided we should use the radiator we have 1214 cm2.

The case is we have a turbo charger that cools by oil and has a water cooling backup system in case it over heats for any reason .
Plus we cannot rely totally on our old radiator because some team members reported overheating of engine sometimes .

So i thought of installing these cells in the water tank ... and make them cool water in case of over heating in engine or turbo (in case cooling water temperature rises above normal for any reason).

So it won't work all the time .... only they start working to save the day when overheating happens ..

I want to hear your opinions

Regards </div></BLOCKQUOTE>

Most teams struggle with cooling issues due to improper ducting, not using a fan, undersizing the radiator or fan, poor engine tuning, or some combination of the above. Your proposed solution sounds like a tremendous over-complication.

I would also advise against running a turbo.

[Aly]

Dear all,

I'd really want to thank you all for this extra useful discussion

@ grantguru :thanks man for these painful facts http://fsae.com/groupee_common/emoticons/icon_biggrin.gif
they really helped as well as hurt http://fsae.com/groupee_common/emoticons/icon_smile.gif

@Zac I'm really sorry for this over-complication
but the case is we are underfunded and we don't have the luxury of ruining anything .
So it's better to be safe instead of losing parts ..

I thought of another thing and I hope no one laughs as it might be a bit childish

Now I want to make an emergency cooling system that works for nearly 20 seconds or something in cases of over heating .... just before failure ...

I thought of placing ""small"" containers containing compressed air . or even a big one .

connected to nozzles that are pointed on the engine (say 3 of them) and another 2 nozzles pointed at the radiator and one pointed at the turbo

when over heating happens
the compressed air should be released .
This large density , high speed flow would cause an excellent boost in the heat transfer process withdrawing large heat amounts in these few seconds .
Thus temperature is retained once again through acceptable limits and an indication is given to show that there has been a problem that caused overheating ....

does this sound realistic ?

[Warpspeed]

No it is not realistic to cool down several litres of water, plus several Kg of aluminium in "just a few seconds". Think about it...

Nothing is going to go into instant terminal meltdown if you have some temperature gauges and carefully watch what they do during initial track testing.

If you DO have an overheating problem, you will see the temperatures rise, and rise, and rise, and you shut it down long before anything is seriously harmed.

[Aly]

thanks tony

[Buckingham]

Why not run a hose from the driver's suit to the radiator so when the engine gets so hot that the driver pisses his pants, the urine will flow through the tube and spray onto the radiator. The evaporative cooling due to the phase change of the urine to urine-vapor will cool the radiator better than blowing compressed air.

What makes you think that storing a finite amount of pressurized air in a tank and blowing it over the radiator is a better idea than using an electric fan to blow an infinite amount of ambient air over the radiator???

[TMichaels]

Aly,
you should consider to use a table in your ECU which links water/oil temperature to the RPM limiter and just restrict the maximum RPM to prevent the engine from overheating. I can tell you from experience that this is quite effective.

[Aly]

@Tobias
thank you very much that's an excellent idea . Hope we can do it ... but we can't do any modifications on our ECU

@Buckingham
compressed air is more dense and has a higher velocity so it would remove more heat .
As for the pissing issue I don't accept this type of humor . It's not funny at all .

[TMichaels]

Well this can be done easily without modifying the ECU. You just need to design a small pcb which reads the current rpm via the camshaft signal for example and which also monitors the engine temperature. The circuit just needs to cut the signal to the ignition coils, if the rpm and the engine temperature are above a certain value.
This could even be done without a mikrocontroller with some discrete and analogue components.

However, I totally support the view of Zac: Get rid of the turbo. It is too complicated for your team, if you are struggling to tackle the problem of proper engine cooling.

If you get the car to drive on its own power at comp than you achieve more than 30% of the teams showing up. That should be a good aim for your team in your situation.

[Warpspeed]

I agree, ditch the turbo it will be a lot more trouble than it is worth.

I still think you are over complicating things by assuming overheating is inevitable, rather than fixing the problem at the source.

I would fit a thicker radiator core and the most powerful shrouded electric radiator cooling fan that would fit into the available space, along with easy to read oil and water temperature gauges.

Deliberately limiting rpm or power in a race car makes no sense at all.
Fix the cooling system rather than add a bunch of other stuff that can (and will) go wrong at the worst possible time.

[Buckingham]

"For the Evo VIII, Mitsubishi has ... A 3-nozzle intercooler jet system sprays water on to the front of the intercooler which helps cool the compressed air going through the intercooler providing more power under hard acceleration."
http://www.supercars.net/cars/2383.html

Spraying water mist onto the radiator would have a significant cooling advantage over spraying air because heat transfers to water faster than it does to air and because the water will undergo phase change to steam, absorbing even more thermal energy (speaking seriously).

Using the driver's urine would simply eliminate the need for an additional storage tank (not speaking seriously).

Agree that your 2nd safety net should be to link water/oil temp to rev limiter. Your first safety net should be to link water/oil temp to the fans for your water/oil cooler.

[TMichaels]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Warpspeed:
Deliberately limiting rpm or power in a race car makes no sense at all. </div></BLOCKQUOTE>
It does as a last resort, if it makes you able to get through the last two laps of Endurance.
I did not propose to limit it all the time.

[Gruntguru]

I need to repeat my original question. Has any of your team studied any thermodynamics? You need to take an analytical approach and many of your questions and "wierd ideas" can be quickly eliminated. For a start, how much air do you need to make a difference to your engine cooling load? Well, air has a specific heat of about 1 kJ/kg.K ie to remove each kJ of heat you will need to raise the temperature of a kg of air by one degree Kelvin (or Celcius). So if your radiator is increasing the temperature of the cooling air flowing through it by 40 deg C, each kg of air flowing through the radiator will remove 40 kJ of heat. A kW is one kJ per second so an aiflow of 1 kg/s will be providing 40 kW of cooling (which is about enough for an engine running at 40 kW output constantly). A kg of air is about 1.2 cubic metres or 1,200 litres. Can you see how much compressed air you would need to make a significant difference - even if you use the compressed air to drag extra air through the radiator?

[Aly]

Yes I've designed the radiator analytically using thermodynamics equations and assumptions .

Those things I'm suggesting are just brainstorming suggestions for overheat cases .

I don't need to withdraw 40KW by this compressed air , it's only something to just remove a little more heat to make you able to run for the last lap or the last 2 laps
yet it's totally impractical ....

I was just thinking of something to act when we are in the endurance and something went wrong and heat is piling up ... something to make us continue ...

We would probably go for Tobias TMichales suggestion it's rather more realistic and effective

regards

[BeunMan]

While this is not my expertise by far (and someone might help here) but here are my ideas:

- Let your ECU inject more fuel (vaporizing and thus removing heat and not burning it all to avoid adding more heat) (DON't know if this will work).
- The energy needed for peltier devices can easily run a high powered fan -&gt; Full radiator fans + fan for engine.
- Turning of the engine at [overheat] temperature.
- Reducing the amount of energy added by some way; limiting rpm, adding less fuel or something else.
- larger water storage (water can absorb some energy), or using bypasses to add 'cool' water if needed.

Serious note:
Check your power (Amperage) budget before turning on a fan: If you disable your controller (e.g. ecu) by drawing too much power (most 12v fans use &gt; 5A including and injectors and other power-hungry devices at high RPM's might overload your power supply) and cool down your engine by turning if off ...

[Rex Chan]

I heard Monash say they ran a separate battery for their thermo fan, if your charging system can't cope (total loss battery-fan). Easiest, simplest backup solution if you designed your cooling system wrong.

Tobias: we do this with the MoTeC PDM - we have "dangerous engine" conditions set up, so that the PDM will turn the engine off if water/oil gets too hot, or oil pressure drops to low (for it's RPM). Of course, you can make it less drastic for comp (vs. track days where I'd prefere the engine to stop if something goes bad), like impose an RPM limit. This actually "saved" us in 2011: last lap of AutoX (Sat) we were in the process of overheating when the PDM shut it all down, thus causing our car to roll over the line. We fixed the offending electric water pump connector (by hard wiring the pump), and had no (engine) problems during enduro {those suspension boys on the other hand...}

Tony: I know some drivers may not like it, but I (person in charge of the engine) does not trust the driver to do anything if the engine is failing. We have a warning light, but it's main job is to let the driver know WHY the engine has shut down on him (i.e. not an accident/ran out of fuel).

Fun fact: I once put in a condition where the car wouldn't run without the GPS (as no GPS signal = less useful data logging, so it's better to fix the GPS before driving any more). Then someone broke the GPS, we removed it, but forgot about the condition, so the car wouldn't rev past 6000RPM...

[Hannes]

Two independent batteries might be difficult with one main switch.

Appart from that issue, I'd never recommend an automated emergency shutdown during racing. Every car is likely going to survife one autocross under any temperature condition and in an endurance I'd push the car until it is completely broken.
Appart from that, sensors brake often or give wrong datas, especially if you're having problems with ground (Bat minus).

2009 our cooling system "exploded" because of complete overheating just after finishing endurance. The engine was broken but we finished the endurance...

[Warpspeed]

If you don't trust your driver to use some judgment and common sense when driving the car, then you really need another driver.

[Rex Chan]

We did have a discussion about this. We decided that any even apart from the last enduro (we get 2 heats in AUstralia), we'd want to protect the engine as much as possible.

We're a 100% new team every year, so our drivers are always inexperienced.

In the event of poor grounding giving wrong data, I'd like to know about it (having an engine shutdown is a pretty good indication). In any case, poor grounds means the data collected won't be very useful, and the engine won't run right anyway.

I can see where you're coming from (real racing), but I like to take the longer term view (i.e. next year's team). Allowing one engine to be damaged because I didn't do my job properly is a bit selfish with regards to the next team who inherits one less engine.

[Fantomas]

I disagree.
Finishing Endurance in FS events is vital for a good overall placing which will probably gain/keep more sponsors outweighing the cost for an engine. Additionally finishing endurance is better for team moral than saving an engine.
This may be different in FSAE-A as you have two endurance heats, but for all other events, it holds.

[kcapitano]

100% agree with Tony.

All this talk about control logic and complex cooling systems, but people seem to be forgetting about the supercomputer they have given full control of their car to (aka. the driver). If you want a system to avoid engine over heats, is simple and guaranteed to work, then how is this:

Put a massive, red indicator light on your dash and couple that to the loudest buzzer you can find. Have the system turn on when your engine reaches critical temp. At that point your "supercomputer" reduces throttle input until his/her ears stop ringing.

As Tony said, if you can't trust your driver to keep the car in one piece then you really do need a new driver. Besides, why can't you trust your driver, but you can trust the people building your circuitry and cooling system?

@ Rex Chan: <BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> We're a 100% new team every year, so our drivers are always inexperienced. </div></BLOCKQUOTE>

There must be someone at your school with karting or some type of competitive driving experience. Find them.