Power source

[Dave Goldsmith]

Hi, I'm trying to design a circuit that I can plug into the wall here in the us (110VAC) and get 12V out that switches polarity every 3 minutes. Also, I need the circuit to have a tunable range of 4A to 9A output, with the constant 12V. I have seen some ideas that involve using a 555 timer for the timing, and I'm thinking an H-Bridge can be used for the polarity switch, but I'm worried that my current requirements will affect the operation of the 555. Any ideas how to design as simply and cheaply as possible would be greatly appreciated.

Thanks,

 

[ronv]

If you hold the voltage constant the only way you can change the current is to change the load.
For example:
If you hook a 12 ohm resistor to your 12 volt supply it will draw 1 amp.
If you hook up 2 of them they will now draw 2 amps.
If you drop the voltage to 6 volts they will draw 1 amp.
But you can't tune the current and leave the voltage constant.

 

[Dave Goldsmith]

Hmm. I've been doing so many circuit problems with independent current sources that they start to seem like magic. I guess I want a tunable voltage then, from 9-18V
thanks for the reply

 

[ronv]

Yes if you want to hold the current steady you want a current source.
What are you trying to

 

[Dave Goldsmith]

I have a bad back, and contrast thermotherapy helps me a lot. I bought 4 peltier TEC's (TEC1-12706) that I want to
be hot for a few minutes, then cold for a few minutes, and repeat for me non stop. From the specs I can find, they seem to be Vmax 18V and I max around 8A, they're sold as 60 watt, but can supposedly handle up to 92 watts it looks like. I was hoping to have the tunability in case they are too hot/cold, or not hot/cold enough, that I could adjust it to be the right temp for me.

 

[Dave Goldsmith]

Also, I'm hoping to have a set of batteries that equate to the desired voltage, so that between having it plugged in it will run on the batteries temporarily

 

[chemelec]

These Modules can get Hot or Cold Very Fast.

But you also should have Opposing Heat and Cold Sinks on them.
It would NOT be Practical to Just put the Module directly on your Skin.

Also, Too Much Heat will damage the Module or BURN YOU.

And for the COLD Side to get COLD, You need a Heat sink on the Hot Side to Dissipate the Pump Heat into the Air.
Or the Cold Side will Soon Not get Hot.
And the Hot Side Will get TOO HOT.

You might read this Article:
http://chemelec.com/Projects/Peltier/Peltier.htm

Want to Talk, I Have Free Phone Service to the USA.

 

[Dave Goldsmith]

I figured I could put an 1/8 inch sheet of steel on the non skin side to hold heat when it's in cold (to me)mode and then it would pump that steel heat back into me when it's in hot (to me)mode. Essentially pumping the same energy in and out of my back

 

[atferrari]

Be prepared to spend lot of $$$ in batteries.

 

[chemelec]

If you plan to use a Battery for all 4 modules, it better be a Car Battery.
That might last a few hours!

A 1/8" Steel Plate, And How big of a area?
But a Flat Steel Plate will Not do much of anything.

The Heat Sink needs to have LOTS of SURFACE AREA.
You also Need a Thermal Grease between the Module and your Heatsink.
At about $20.00 Each, you Don`t want to Burn them out in 2 Minutes of use.

DAVE, If you Want to Talk about this, I Have Free Phone Service to the USA.
Just Email me your Phone Number.

 

[Dave Goldsmith]

The tec's were about $2 each. They're about 3x3cm, so with four in a square configuration I need about 9in^2 steel area. It won't really be a heat sink, but more of a heat tank. I will definitely put thermal compound between the tecs and the steel, and then hot glue around the sides to hold it to the steel. I hadn't done any math on the battery costs yet. I did a little last night and found the cheapest was around 7.25cents/kJ. At 400 W, that means it's about $1.74/minute to run the thing and with 4 of these https://www.dx.com/p/protected-1865...-ion-batteries-yellow-pair-82143#.VMzNny4YFQI I can get about 15 minutes for $30. Thanks for all the info. I still need to figure out the cheapest, easy, reliable timer and h-bridge type polarity switch. I think I might use a laptop power supply from good will.

 

[KeepItSimpleStupid]

You can PWM (Variable duty cycle) the module. What kills them is when you exceed the specified delta t. Usually about 40 to 60 Centigrade.
Condensation is possible. Totally forget batteries.

I didn't know there were IC's for this: https://www.analog.com/static/imported-files/data_sheets/adn8830.pdf

https://tetech.com/product-category/temperature-controllers/

 

[chemelec]

The tec's were about $2 each.
That is an UNREAL. Price.
The Prices I found on the Internet is $20.00 each in Unit Pricing.

The Metal plate will NOT Act as a Heat Storage very well.
Hot Melt Glue WILL MELT with the Heat.
Besides it WON'T Hold the TE, Solid Enough to the Metal, Even when COLD.
And the metal will get Quite HOT in 15 Seconds.

The Timer better be Your Finger on a Switch or you will either get Frost Bit or BURNT.

 

[Dave Goldsmith]

KeepItSimpleStupid thanks for the info on PWM, I just assumed that it would work fine, but in hindsight I can imagine how that could be a problem. In my experience with these tec's so far, I doubt temperature controllers will be helpful. They are reasonable to hold in your hand, hot or cold side, at 12V (from an agilent E36xx, .55A max on the 20V channel, set to 12). I wonder if they were only holdable because of the current limitation of the power supply?

chemelec, thanks for the info as well, https://www.amazon.com/Vktech-TEC1-...=2025&creative=165953&creativeASIN=B00IKDL22O
these are the tecs I bought, I underquoted a bit by accident, I bought a 5 pack for 18 shipped, so it was closer to $4 each. That link is $2.50 each
The metal plate will be surrounded by fabric, with a wad of kerlix behind the plate (opposite me and the tecs) to help hold the tec's securely against my back. I think that the plate being so insulated should help it work effectively as a heat tank. If you have some reason to think otherwise I would appreciate knowing.

As far as the glue melting and it being very hot in general, I held one in my hand from the agilent 12v I mentioned above. It definetly wasn't hot enough to melt glue, but that may have been due to current limitations of the power supply.

Do you have recommendations for a better way to secure the tec's to the metal plate (with the thermal compound in between)?

Also, I haven't given you my number yet because it seemed like you were fishing for phone numbers. Not because I don't appreciate your help with this, which I do. Thanks again, Dave

 

[Dave Goldsmith]

I think I saw that the deltaT on these was only 25C, but I can't find that again anywhere

 

[chemelec]

Hi Dave, According to the Data Sheet for this TEC1-12706:
If Hot side is 25 Degrees C, DT Maximum is 66 Degrees C.
If Hot Side is 50 Degrees C, DT Maximum is 75 Degrees C.

Some how I must have looked at a Different TE Module.
This one is a Cheap Chinese Knock off.
But it does look Fairly Good.

If you look at my Website, Main Page, you will see MY Phone Number, Just below my Email address.
I Hate Typing and Phone Conversations are MUCH Quicker for Numerous Questions and Answers.
I Do Not Save, Collect or Pass Along Phone Numbers to Anyone else.
Your Choice to talk or not.

Yes, Being current limited At Only .55 Amps, you can probably Hold it OK.

If your Hot Side STORES the HEAT, it will deter your Cold side from getting very Cold.

As Above: If Hot Side is at 50C, DT is 75C MAXIMUM.
More Likely it will be Considerably LESS than that 75C.

Take Care...Gary

 

[KeepItSimpleStupid]

I used TEC's at work with "controllers". However the "controllers" were very dumb. Just a bipolar power supply and a readout designed by a TEC manufacturer, Cambion.
They are very slow to respond. The largest module was about 4" x 8" and was air cooled in the back. The others we used were water cooled on one side. I did use one module in an air to air mode cooling an enclosure.

 

[ronv]

Yes, the reason yours are not very hot and very cold is because you don't have enough current.
They are quite inefficient in "making" cold, but very good at "making" heat. A nice way of saying they are inefficient.
Usually the hot side heat sink is much larger than the cold side to make up for this. It's 12 volts at 5 amps so think of it as a 60 watt light bulb - pretty hot.
Often the way they are set up is with a cold plate, the Pelletier, a small spacer (say 1/4 inch) and a good sized heat sink. In between the 2 you can use insulation (Styrofoam) then clamp the 2 sinks together with the device in the middle. You need to insulate the screws with a sleeve so they don't transfer heat from one side to the other. It will become a little bulky.
Edit: add link

https://engin1000.pbworks.com/w/page/23294658/Drew Plichta, James Develin Cooling Knee Sleeve

 

[KeepItSimpleStupid]

Yep, what he said.

You might be better served, with two water baths. Hot and cold. Now just alternate the hot and cold water baths through your circulating loop.

I know it's "less portable", but I think it would work much better. If you wanted colder, then use an antifreeze solution.

e.g. https://waterbaths.com/products/refrigerated_heating_circulators.html

Just to see if it works, you could try your hot and cold water taps. That's like 55F and 140F. Want a little colder, put some of it in an ice bath.
With a circulator, you would get mostly regulated temperatures and they would not be out of control.

 

[KeepItSimpleStupid]

3 minutes is a very poor time constant for the TEC.