Sharp 8A Solid State Relay and DC output?
jmspaggi
Posts: 629
Hi,
Does anyone know if I can command DC device with a Sharp 8A Solid State Relay? In the documentation, they say it's perfect for AC, but is it also working for DC?
http://www.parallax.com/StoreSearchResults/tabid/768/txtSearch/sink/List/0/SortField/4/ProductID/657/Default.aspx
My goal is to command +12V DC directly from my propeller (if possible).
Thanks,
JM
Does anyone know if I can command DC device with a Sharp 8A Solid State Relay? In the documentation, they say it's perfect for AC, but is it also working for DC?
http://www.parallax.com/StoreSearchResults/tabid/768/txtSearch/sink/List/0/SortField/4/ProductID/657/Default.aspx
My goal is to command +12V DC directly from my propeller (if possible).
Thanks,
JM
Comments
Thanks for your reply. So I will try it. In the documentation, they are talking about snubber circuit to add. Have you done that too? Or simply connected the relay directly to the controller?
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
Careful, it is a SOLID STATE relay, not a mechanical kind. Mike seems to be thinking a mechanical device is involved.
This is rated for 240VAC, not for DC on the output side. You may very well find yourself with nothing working.
If you want to control 5Amps of 12VDC, try a darlington TIP120 with a 4.7K pull down between the base and ground. Also you may want a small resistor, like 100ohms from the Propeller to Base to protect the Propeller.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Thanks for your post. I will try with a TIP120. The goal is to control the heater fan in my car to regulate the temperature. So I think 5Amps is way enought.
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
http://www.sullivan-county.com/ele/triacs2.htm
Some links that have opto-isolator information:
http://www.epanorama.net/circuits/parallel_output.html
http://www.edaboard.com/ftopic158099.html
On the first link you sent, they are talking about MOC3011. So should I put the MOC3011 between the Propeller and the TIP120? Will not the 100 resistor between the TIP120 and the propeller protect the controller?
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
You'd want to look at some of the other ones for use with a DC load. Common ones are 4N25 and H11A1
I wouldn't feel comfortable just using a 100ohm resistor to protect the Propeller for the type of load you want to control.
As I recall I think there may be some examples in the Nuts'N'Volts column for the stamp that use an opto isolator:
http://www.parallax.com/tabid/272/Default.aspx
A couple other links:
http://en.wikipedia.org/wiki/Opto-isolator
http://www.radio-electronics.com/info/data/semicond/phototransistor/photo-optocoupler-optoisolator.php
They usually have a Triac as the pass device.
If you were to turn it on it would not turn off unless you interupt the DC somewhere else.
The best method would be to use a DC Solid State Relay.
They are isolated and easy to use.
Another option would be to use a ULN2803 chip. It has 8 outputs and can handle 500 ma each, and you can parallel them.
I use them directly off the Stamps I/O pins with a 10k pullup(to +5) resistor.
They have a built in flyback diode, so you can directly drive relays and small inductive loads directly.
I dont have a pdf converter for my schematic program at home but ill try to remember to post one tomorrow when I get to work.
Good Luck
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Alan Bradford ·N1YMQ
Plasma Technologies
Canaan NH 03741
www.plasmatechnologies.com
You could use one channel of a hex driver chip to isolate the Propeller as well. It would fry rather than the I/O on the Propeller if anything came back at it (like a direct short or 12VDC).
But, in an automotive environment, the power supply side of the Propeller is where you need to prevent spikes of up to 100VDC from creeping in. LM2940-3.3 regulators seem to work very well at trapping this kind of trash. Also 30V spike suppressors on the power in can precondition the power line.
One can get carried away with protection and junk up a reasonably good design to the point that it becomes difficult to know what's wrong. Keep it simple and go with an opto-isolator, a resistor, or a hex driver chip - certainly NOT all 3.
The TIP120 is rather for 5 amps of output and up to 60 volts. If you don't need that much power, you could use a smaller darlington or just a plain old BJT.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
I like the opto-coupling because I already burned few Javelins and propellers with my tests [noparse];)[/noparse] The hex driver, I don't really know how it works, so I will probably go with something like 4N25 and then the TIP120 after it.
Can I use 3.3V coming from the opto on TIP120 base, and 12V on the collector?
Since I have to command 3 TIP120, do you know if there is some 4N25 with 3 circuits? Or will I have to put 3 of them in a raw?
BTW, what's a BJT?
Thanks,
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
A BJT is a conventional transistor - 'Bipolar Junction Transistor'.
Yes, the maximum Emitter-Base voltage is 5 volts so 3.3 volts isn't an issue. But you may want to provide a current limiting resistor as you don't need unlimited current to drive the TIP120 to saturation (full ON). 2ma seem minimum, but according to the PDF, 20ma will prove 4amps of power at 5 volts. You need much more for a full 5 amps at 12 volts.
I have no idea how many amps you need, but 5 amps are the maximum available. For that 4 amps at 5 volts, a limiting resistors of 165 ohms is used.
3.3volts/.020amps = 165 ohm current limiting resistor (max of 20ma)
Here we go on some calculation from the PDFs to size resistors to protect all and everything...................................
Can the opto-isolator deliver 20ma or more? Yes, up to 150ma, but the TIP120 will only take 120ma of base current max - after that you get smoke.
Actually this is better than a hex driver (LS7404) which may output only about 8ma per channel. (you could combine more channels as required, but it looks like 48ma tops because a hex driver only has 6 channels).
So if you don't have enough power to the fans, use a smaller resistor, but make sure you protect the TIP120 base. 3.3volts/.120amps = 27.5 ohms is the absolute smallest current limiting resistor you should use.
Be careful to allow some leeway. Try a 3.3 for a max of 100ma to give you some leeway as resistors vary %5 or so. 27.5 ohms is rather odd anyway. And that should be a 1/2 watt resistor as 1/4 watt might burn up. 3.3 volts x .1 amps = .33 watts (greater than 1/4 watt, less than 1/2 watt).
The LED on the input is rated at 1.15volts max and 10ma, but you have 3.3volts supply and possilbly 30ma.. 3.3volts/.010 = 330 ohms would protect the LED.
But the Propeller should be protected as well. Can it output 10ma? or do you want to output less? Less is always better as things run cooler.
Let's limit this to 5 ma. 3.3volts x .005amps = 660 ohms between the 4N25 and the Propeller pin. This will make the 3.3 volts unimportant as there will be a voltage drop to the opto-isolator and it will be fine. And the 5ma is safer for the Propeller. The LED will be safe, but a little dim.
Someone might jump in and correct my maths. The output for the opto-isolator may only be 2.6 volts (due to a diode voltage drop of 0.7 in the photo-transistor). In that case 2.7 volts x .100 amps = 27 ohms, not 33 ohms.
If you still use 33 ohms, you get less overall available amps. 2.7 volts / 33 ohms = .81amps or 88 mill-amps. Still that may be adequate and is certainly a safe size to start with. If everything works, fine. If the fans need more power, use a smaller resistor. 2.6 volts/ .120 amps = 22 ohms. That might be a more realistic absolute safe limit and that still is a 1/2 watt resistor.
Hopefully you can follow all this. As you can see there is a lot of back and forth in calculation and room for error needs to be included.
And as I said in the beginning, everything depends on how many amps you need to deliver at +12 V (automotive really can be up to 14.5 volts due to battery charging). If you only need 1 or 2 amps, I think everything will pretty much work without change and last a long time. But if you really need 5amps, it becomes important to double check everything and maybe redesign.
In sum, use a 1/4 watt 680 ohm resistor between the Propeller and the 4N25. And use a 1/2 watt 33 ohm resistor between the 4N25 and the TIP120.
If you have enough power for the fans, fine. If you don't ..... well nothing got destroyed and you can look at where to redesign. Likely the 33 ohm can be moved down to as little as 22 ohm. But going halfway to about 27.5 is better.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 7/28/2010 7:40:00 PM GMT
Thanks a lot for all those details.
I think I understand the principle and the numbers. But I will not try to modify them [noparse];)[/noparse]
The fan has 3 states. From slow to fast. So if 4A are not enough, I will just try to lower the speed. Then if the prototype is working and I really need more, I will read this post again and again.
I ordered the TIP120 today, I will get the 4N26 (there is no 4N25, hope it will work the same way) tomorrow.
I will keep you posted of the result [noparse];)[/noparse]
Thanks,
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
After thinking about the LED resistor, you might consider providing 10ma to it and get full power available through the opto-isolator if overall output is poor.
5ma to the LED may not get enough final output amps from the TIP120. That would mean a 330ohm resistor instead of the 660ohm one. But if you have any doubts, you could go something in between 390ohms, 470ohms, 510ohms, and so on.
Hopefully you will begin to see that you must limit current or things will burn up. Also, that you need to build up stages to get a micro-controller to provide a low of power in the real world.
I now see that the TIP120 works well with a direct connect to the Propeller for 5 volts (20ma from the Propeller will provide 4amps at 5 volts), but you really need the extra stage to boost the output to 100ma for higher voltage. Still the 100ma may be overkill as the TIP120 is good for 60Volts at 5amps.
It is all those darned curves and charts in the PDFs that take time to comprehend. And you still have to estimate and do trial and error. I have a BasicStamp here that is reliably running 12 volt relays (about 60ma draw) from a TIP120 that is restricted to just 1 ma. I have been running it on a 1 second cycle for months now with any failure.
It would help is you measure the full on current of the fans. Then we could stop all this guessing.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
I took some values and here is what I have.
Full speed: 5.2A
2/3 of the speed: 3.2A
1/3 of the speed: 2.1A
On the motor, I have 4 connectors. One for the ground, and one per speed. So Full speed is to much. I will simply connect to 2/3 of the speed which is enought.
So based on those values, what should I do? If 3.2A is still to much, I can go with the 2.1A option, but I will prefer the 3.2A.
Thanks,
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
First off, if you are running the TIP120 at 5amps, you need serious heat sinks or it will burn up.
Second, by your resistors you can limit the over all maximum value to something less than 5amps output, and/or you can further limit the output by PWM.
I suspect that 2/3 speed is the safest way to go with what we have talked about already.
Darlingtons say they has a gain of 1000, which would mean 1 ma in puts out 1 amp; 5ma in puts out 5amps. But it isn't that precise. It is a curve and there are different curves for different operating voltages. I look at the PDF, and there is no clear information for above 5 volts. Operating at 12volts is okay, but guess work is involved in sizing things.
The main thing is to get the opto-isolator right so as not to damage the Propeller. The second thing is limit any possible damage to the TIP120, if it might come to that. You certainly don't want to damage the fan.
But you can seek out another TIP1XX, the TIP130 is rated for 8amp and will allow you some headroom against failure. The maximum base current is 300ma, so that is above and beyond the 2N26 output. You will drive it at 150ma tops using 3.3/.150=22 ohm 1/2 watt, or something safer.
Since this is a rather ambitious amount of power to drive, you may want to Google for other TIP130 circuit examples and try to find what is wrong or overlooked.
in my suggestions. The motor should have a flyback diode on it to eliminate spikes. The internal diodes in the darlington are helpful, but these additional diode should enhance long life
If you want to drive the TIP130 over 150ma, you will have to add another stage or find an opto-isolator that has greater output (I believe some have a darlington in them for just that purpose), but I saw someone claim that the TIP130 has a gain of 2500 @ 12v @ 4amps. So it is a trial and error process with some careful thought.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 7/29/2010 2:39:26 PM GMT
I already ordered the TIP120, so I will go with them and with the 3.2A option. I don't really need it to be at max throttle.
So are you said in a previous post, I will use a 1/4 watt 680 ohm resistor between the Propeller and the 4N26, and a 1/2 watt 33 ohm resistor between the 4N26 and the TIP120. That way, everthing should be protected as it should, and the TIP120 should be able to deliver enought power to the fan.
Correct?
The more I read your messages, the more I understand, but I think I still have a lot to understand [noparse];)[/noparse]
Thanks,
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
Then consider changing the 33 ohm to progressively smaller values - but I hope that isn't necessary.
Don't forget the heat sink!! The TIP120 could dump 70watts of heat. And, you really need the TIP130 to get full performance.
Just think what it is like to hold a lit 25 watt light bulb in your hand and you will get some idea of what Watts mean. Watts always has a tangible heat equivalent. A lit 60 watt bulb left on a mattress could burn a house down. I've seen it almost happen. 120 watts will blister your skin in a few moments.
I suppose the ideal way to do this is to build a breadboard and play with the components until you are satisfied. Alternatively, if you can understand SPICE, use the software. You may notice that 12 volts in, drops down to 10.6 volts to the motor. Darlingtons have two diode voltage drops (2 x 0.7 volts = 1.4 volts), but automotive motors generally are insensitive to such as the battery drops low when one starts the engine and output runs as high at 14.5 volts when charging the battery.
It is those diode voltage drops that create the heat in the transistors.
Of course, you could graduate up to a TIP140 and 10amps. But that is yet another story - bigger toys for bigger boys. Costs go up accordingly.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 7/29/2010 3:43:13 PM GMT
You need the plug compatible 4N33, which will supply up to 500ma. (maybe a 4N29 will just barely do (at 100ma), but I am rereading everything).
And I wonder where you are getting all that 3.3volt power from. Check your 3.3 volt for adequate capacity.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 7/31/2010 4:50:11 PM GMT
The 4N33 is available in my local store, and I have not yet received the TIP120. So I will try that soon with the right pieces.
Regarding the 3.3, I will use a LM2937ET-3.3 V which can provide up to 500ma. So I should be correct, no? Else, I bought 3, so I will put 2 in //?
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
I am happy to hear the 4n33 is not a problem to acquire. I was quite surprised by getting a bad information sheet.
I assume the +12v is coming from automotive mains and current demands are no problem. But consider including appropriate fuses.
In general, voltage regulators cannot simply be put in parallel at the same voltage to get more output, they oscillate and become unstable, burn up.
There is an exception if one is putting out something like 0.2 volts higher than the other, but that is only useful for creating a backup battery circuit as one and only one is outputting at a time. The one set for a lower voltage remains off until the higher one shuts down.
At this point, I don't know if you are using just one TIP120 or several at the same time, so I don't have a real idea of how big a power supply you require. I had imagined that you wanted to use 3 TIP1xx's to turn on the fan at different speeds, but one at a time. If that is still the case, one should be a TIP130 for the 5.2amp for full speed; but the 2/3 and 1/3 can easily be handled by the TIP120. All can use the 4n33. Since the fan won't take more current that required, I think the TIP120s can work with the same current limiting values for 1/3 and 2/3 speed.
If just the output side of the opto-isolators are powered by the 3.3 regulator (not the Propeller), that would be 100ma+100ma+250ma= 450ma. It seems good, even if you turn all 3 on at the same time. And the output side of the opto-isolator would be on a separate circuit from the Propeller and input side. I am not sure how you are using the Propeller, so I would get it is using a bit, say 400ma of power. That would mean it should have an independent power supply.
Frankly, the whole set up is easier to do by just having a 2n2222 drive a TIP120 or TIP130; or having a 2N7000 drive a TIP120 or TIP130. One might even supply +5 (but no higher - the TIP's base is rated at 5 volts max) for the interim stage, but no formal isolation -- just current limiting resistors on the bases/gates of the transistors/MOSfet.
Both the 2n2222 an 2n7000 output quite a bit - the 2N2222 can drive 500ma tops , while the 2n7000 can drive about 200ma tops. So I think you see the opto-isolator is more limiting, as well as being less responsive, and a slight bit more wiring.
Before you assemble everything, tell me what you have and we will run through this one more time. I am willing to breadboard a 4N33 and TIP120 here and see what happens. I am particularly interested in how it handles being driven to maximum output from a few miiliamps.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/1/2010 7:19:47 AM GMT
Too much writing and no schematic. Here is diagram that may cut through all the talk. The far side of the opto-isolator appears to be taking 12vdc, but letting much less get to the TIP120 by using a voltage divider. No need for a lot of thought about power supplies and regulators.
http://markshauntedgarage.com/halloween/techinfo/spider.php
The above website uses an 4N33 mated to a TIP120 quite nicely. No extra resistors that I thought might be needed.
I suppose the we should check the math though to optimize output. Here is a start.
Personally, I dislike both the 1K resistors. The one on the input seems a bit too conservative, lowering output and heavily diminishing the LED's glow. We had a very conservative 660ohms before and I suspect that 470ohms would do fine. I don't think this guy is getting high amps out of his circuit, but reducing that LED resistor could burn up the base of the TIP120. I'd prefer to have the opto-isolator and the TIP120 both running at about 80% their limits. After all that is about 4amps.
The 1k on the upper side the output of the opto-isolator may allow as much at 150ma to drive the Darlingtion (assume 15VDC peak from automotive), 1.5K ohms would seem to be closer to limiting the base to a safe 100ma, but I also don't see how the base voltage is limited to 5Volts or less. I guess the 10k is fine as a pull down. Resistors still have to be looked at for wattage rating.
Further I do wonder about automotive power spikes (peaking at 100VDC) doing nasty things to the base of the TIP120. There are low voltage Metal Oxide Varistors that could trap them. Some people do claim that certain voltage regulators do prevent the voltage spikes from getting in. So maybe you prefer to drive the TIP120 from an independent 3.3v regulator instead. Still both that and the Propeller's regulators are best if protected by a low voltage MOV (something in the 20-30 volt range).
But a schematic is still needed, for sure. I will try to buy a 4N33 tomorrow, for a bench test. All I have is 4N25s.
I tried to locate some SPICE simulation software for this, buy so far nothing I locate for free has the components we are using.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/5/2010 12:44:30 PM GMT
The opto-isolator should be a 4N33, but I can't seem to make the CAD program change the name. It keeps reverting to 4N29.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/1/2010 6:07:05 PM GMT
Regarding the fan, speed 3 is high (5.5A) and speed 1 is to slow. So I think I will only take the speed 2.
I have also +5V available, is that beter to use that for the TIP120 instead of the +3.3V? Or quite the same result?
JM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Linux? There is worst, but it's more expensive.
I imagine we are both a bit fed up with PDF sheets and calculations at this point. The realities are starting to sink in.
1. A lot of the suggested circuits on the web either don't provide enough information (apparently the real market for TIP120s is in driving high amp steppers and companies don't want to publish their circuitry). I found the Google images show copies again and again of the same circuits, but there have been very few independent designs - over many, many years. That leads me to greatly distrust the one and only TIP120/4N33 design posted - it may work well with VDD at 5volts, but flop at 12volts. He doesn't say what the VDD for his steppers are - very sloppy. And he doesn't say how many amps he is driving.
Many, many of the forums on electronics are merely repeating what others have said elsewhere - we seem to all be search engine junkies.
2. Using this in an automotive environment adds a layer of problems due to spikes up to 100VDC. (I got some Metal Oxide Varistors today - rated at 23VAC/30VDC to protect both the voltage regulators and the TIP120). Some people believe that all transistors in an automotive environment should be rated at 100VDC, the TIP120 is only 60VDC - but there is it's cousin the TIP122 that is the 100V version. Should the TIP120 fail in actual use, change to the TIP122. Similarly the TIP130 has a cousin the TIP132. Of course, these cost more money.
3. Building and testing is the only way to really determine a good, rock solid 14.5VDC (max Automotive Volts) 4AMP output (about the safe limit for the TIP120).
So what next?
A. I may be all wrong about needing 100ma to drive the TIP120. (estimates have a way of screwing up) So I got a 1K 1/2 pot to determine what value resistor and how many milliamps are actually needed. I'll feed regulated 3.3volts or 5.0 volts into the base and turn up the pot until the TIP120 goes on with a 3 amp load (two 8 ohm 20watt resistors in parallel). Then I will bring the load up to 4amps and see what the differences are.
B. High amperage bench supplies at 14-15VDC (and higher are expensive), so I am using two gel cells - an 8VDC and a 6VDC in series (very handy alternative, but a car battery or a motorcycle battery will do nicely as well - don't forget to fuse this at 5amps as you can get way lot more.) Please remember that automotive charging peaks ideally at 14.5VDC, building to 12VDC (because everyone says automotive is 12VDC) is likely to mess up your design - 15VDC max and 10VDC minimum is more realistic, with up to 100Volt spikes.
C. I was able to buy a 4N33 opto-isolator, but I am going to build in stages and the opto-isolator will go in last.
(Why so? Darlingtons have two diode voltage drops internally - that is 0.7 + 0.7 = 1.4 volts. The 4n33 is an Darlington going into another Darlington, the TIP120. We could be dropping 2.8 volts of the 3.3volts and not have much luck turning on the TIP120 with the 0.5volts that is remaining).
But don't worry. There are alternatives - power the 4n33 with 5VDC; or drive the TIP120 with a 2n2222a which has only 1 diode drop (0.7volts); or drive the TIP120 with a 2n700X that has no diode drop. Both the 2n2222a and the 2n7000 can merely have a current limiting resistor between the Propeller and them to protect against excessive current drain.
So.... No more estimates of what is what. I will just have to get buys and report back what is or is not working. I'd still like to eventually use a TIP130, or TIP132 to get you all the power you wanted. But that means another bench test.
First I am breadboarding the TIP120 to run stable when at least 3amps is steady on full.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/5/2010 12:47:13 PM GMT
My load is now three 20 watt 12ohm resistors in parallel as I got caught assuming 12VDC when shopping, rather than near 15VDC. I can only easily get 20watt resistors for a test load. 8 ohms is fine at 12 volts, but will burn up at 15 volts.
So tomorrow, I may have some info about actual testing. The packaging for the TIP120 certainly doesn't look substantial enough for 5 amps, especially the leads.
15volts/12ohms = 1.25amps load; 15 volts x 1.25 amps = 18.75 watts
15volts/8 ohms = 1.875 amps load; 15 volts x 1.875 amps = 28.125 watts, way too much.
And of course, the thinking automotive is 12 VDC, not 15VDC max does the following:
12volts/8 ohms = 1.5 amps load; 12 volts x 1.5 amps = 18 watts, looks fine but the voltage is wrong.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/2/2010 4:29:40 PM GMT
I am somewhere in the region of 3.2amps steady on at 12volts output and I am fuse at 4amps, so I know I've not gotten that high.
Let's review some math.
My input.
5volts/34ohms = 147ma - I appear to be a bit 'over the safe limit' of 120ma. But I've not yet destroyed the TIP120.
An estimate of power output.
1. My 2watt 12 volt light bulb -- 2watts/12volts = 0.16 vamps
2. My three 12 ohm 20watt resistors with each providing a load of 1 amp 12volt/12ohm = 1amp
3. The total is 3.16 amps.
I could, get a 50 ohm resistor to limit current to 100ma (or dig up a 41.66 ohm for 120ma) and see what the effect is. But I am going to simply move on to using the 3.3 regulator and see what the 34ohm resistor does with that. I don't want to run to the store for just a 42 1/2 watt ohm resistor.
Time to recharge the batteries as well. They are rated at 4Ah, but I can see the voltage drop as I run the tests. I also need to compare battery voltage to output under load voltage to see what is really occurring at a 3.2am load. There are voltage drops and that is why many prefer power MOSfets to Darlingtons. If you are also doing a bench test, be wary of the heat and use a good heat sink on the TIP120.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/3/2010 8:12:29 AM GMT
What's with this hundreds of milliamps of base current stuff?· The TIP120 has an h_FE of 1000 (guaranteed).
Stress testing prior to a real installation.
I'd rather give advice based on successful construction than muck around with PDFs and see someone else get frustrated by reality. Controlling near 5amps with the TIP120 is not really well documented on the internet, not in the PDF. Besides, from what I have been led to understand, full saturation actually might be best for the transistor in terms of destructive heat.
Gain is both a set of curves and a guide line. I am not sure where you get the '1000 guaranteed' from. I see on the PDF, a gain of 1000 at 4 volts supply for 5 amps, not 15 volts. But still, the 5 amp output seems limited to about 20 volts and any higher voltage degrades the amps output. Since I am near the margins of its steady state output and 120ma is the indicated maximum base, I figured 100ma was a useful starting point to determine what actually is required to get 5amps of output in an automotive environment with charging peaks supposed to be 14.5volts.
Please feel free to point out what I've overlooked.
At this point, it seems that 100ma appears needed to get 3.3volts at the base to fully provide power near to 4 amps at 15VDC in a steady on state. I am using a set of 1/2 watt pots, to adjust the current to the base and watch the light for change in brightness. Admittedly, I may require less to drive to saturation, but I'll explore that after I get the opto-isolator working (or something else between the TIP120 and the uC to get that 100ma).
Most recently, I tried to insert a 4N33 for digital control, but have made a mess of things (may have destroyed one of the components or all of them - blew the 4amp fuse). Right now I am rebuilding.
IN sum, we are just having fun and gaining some useful insight into how to test component before one connects them to a micro-controller. We started with absolutely the wrong component for this task, a solid-state relay. See the beginning of the thread.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/3/2010 2:57:06 PM GMT
But now with a new +5 regulator and a new 4N33, I have the TIP120 working, though I am unsure about output in this configuration as I am only using the 8V cell right now.
The TIP120 is still working fine from either 3.3volts or 5.0volts at up to 120ma under a 3.2amp load (or even nearly a 4.0amp load) at near 15 VDC, no apparent damage.
So we do have something in the way of a functional circuit. BTW, I have omitted the 3.3K or 10K pull down resistor in all my set ups as the TIP120 has an internal pull-down of about 8K.
The TIP120 and the bank of load resistors are all toasty hot
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/3/2010 4:53:52 PM GMT
10ma at 5 volts to the opto-isolator turns it all on, when removed it all goes off - no inversion of logic.
So we have 'proof of concept'. A micro-controller can drive above 4.5 amps in an automotive setting with opto-isolation.
I want to explore a few refinements, recharge the batteries, and do tests with a 3.3v regulator as well, but the TIP120 may require at least 2.5V to the base and the 4n33 may only provide 2.2V due to the internal Darlington pair.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Ain't gadetry a wonderful thing?
aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan
Post Edited (Loopy Byteloose) : 8/3/2010 5:53:18 PM GMT