Shop OBEX P1 Docs P2 Docs Learn Events
Problem with circuit, part failing — Parallax Forums

Problem with circuit, part failing

metron9metron9 Posts: 1,100
edited 2008-09-18 23:31 in General Discussion
I have a circuit that is used for 6V cars that powers a 12V Tach. The Dc-dc converter is failing and i don't know why. Perhaps the grounds tied together? I did just realize the dc-dc part uses a isolated tiny transformer.

Does anyone see a problem with the circuit? I have bench tested it at up to 12V and the polyfuse does its job shutting it down before it can overheat.

http://www.cd4power.com/data/power/ncl/kdc_nke.pdf

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Think Inside the box first and if that doesn't work..
Re-arrange what's inside the box then...
Think outside the BOX!

Post Edited (metron9) : 9/17/2008 11:59:49 PM GMT
5TO12vBOOST.png
1030 x 473 - 6K
DSC06321.JPG
1600 x 1200 - 768K
DSC06320.JPG
1600 x 1200 - 796K

Comments

  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-09-18 00:06
    "Failing".· Insufficient data.

    Rated output current 83mA.

    Does it "fail" with a load other than this tachometer?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-09-18 00:15
    Which zener are you using? With the 200mA polyfuse, you need one that's rated for at least 1W, such as a 1N4733, a 1N3826, or a 1N5918A.

    A much better choice would be to use an LM2940-5.0 regulator ahead of the DC-DC converter. They have a low dropout voltage and were designed for automotive use.

    -Phil

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    'Still some PropSTICK Kit bare PCBs left!
  • metron9metron9 Posts: 1,100
    edited 2008-09-18 00:55
    The tach only draws 30mA max.

    When I short the output through a 100ohm resistor it only shows 80mA through the resistor and drops to around 40 as the polyfuse heats up
    Shorting the output fully initially draws 250mA from a 7V supply and drops as expected as the poly fuse heats up
    Still works fine after shorting as a matter of fact the one he gave me he said did not work and it works just fine.
    I think I will take all the parts he has that he got back and test them, perhaps it is operator error.

    By the wau the RED input wire is not pn the picture above. The white wire is the 12V output.

    I am making some assumptions that the car voltage is at least 7V while running due to the alternator, the part runs stable from 7V up to almost 11 volts, past 11 volts the 200 ohm poly fuse starts to heat up and lowers the voltage.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Think Inside the box first and if that doesn't work..
    Re-arrange what's inside the box then...
    Think outside the BOX!
  • metron9metron9 Posts: 1,100
    edited 2008-09-18 01:06
    The Zener is a surface mount part 5.1V 3000mW

    http://parts.digikey.com/1/parts/421798-diode-zener-3w-5-1v-smb-1smb5918bt3g.html

    However the part on the board says AG on it and is not the same part, it is a ltiile narrower than the above part so I will have to find out what he bought for parts.

    The maximum though should be only 200mA and at 10.5V it is running at 100mA

    I am running it now and it is about 125 F while the polyfuse is at 93F thats at 10.5V and I hope a 6V alternator connected to a battery is below that.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Think Inside the box first and if that doesn't work..
    Re-arrange what's inside the box then...
    Think outside the BOX!
  • metron9metron9 Posts: 1,100
    edited 2008-09-18 07:36
    Trying to figure out what the three black parts are across the top. What makes the square wave? I don't have a good enough magnifying glass to see the little part in the center but I think that is just a resistor.

    Pin left to right is -in +in -out +out· and indeed the grounds are not tied together so I think that may have something to do with not working parts at some point but it seems to function fine with the grounds together. I will try and see if there is any difference using just one of regulators and then with and without the grounds tied together.


    Could the oscillation be from the current through the toroidal inductor so it generates its own pulse like the jewel thief circuit?
    ·

    I removed the inductor so I could see what was under it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Think Inside the box first and if that doesn't work..
    Re-arrange what's inside the box then...
    Think outside the BOX!

    Post Edited (metron9) : 9/18/2008 7:45:38 AM GMT
    DSC06323.JPG
    1600 x 1200 - 875K
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-09-18 07:59
    Wow! It's amazing what crappy soldering companies get by with when they encapsulate their circuits. How did you manage to extract the PCB from the potting?

    -Phil

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    'Still some PropSTICK Kit bare PCBs left!
  • metron9metron9 Posts: 1,100
    edited 2008-09-18 12:41
    I used Google and searched for removing cured epoxy resin. Came up with Acrylic Adhesive· Methylene chloride. It melts some of the other components insides as well. Now what do you think those black parts are "TG3" I think is on two of them. I want to simulate the circuit on my switchedcadIII From Linear Technology, if you have not tried that free spice model program you should download it, they have hundreds of parts you can model and any circuit you can come up with down to nanoseconds on the plots. Here is my boost circuit I plan to use with my Luxeon fansink project. Using a 5V supply and a 12V fansink combo.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Think Inside the box first and if that doesn't work..
    Re-arrange what's inside the box then...
    Think outside the BOX!
    boostplot.png
    873 x 793 - 22K
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-09-18 12:42
    I say it's OK, in this case, to connect "-in" to "-out".· There are situations when that would be inappropriate.

    I disagree with the zener diode "regulator" thing you're doing.· You have a 10Ω resistor in series with your load.· With 80mA out, your boost module is taking nearly 250mA in; no achievement in efficieny there.· Brash assumption, 40mA out means 125mA in; 125mA x 10Ω = 1.25V, 6V - 1.25V = 4.75V, getting marginal.· The range is 4.5 - 5.5V.· When it is further subjected to environemntal heat then that takes us further into Derating Land.
    boostreg.JPG
    321 x 415 - 15K
  • metron9metron9 Posts: 1,100
    edited 2008-09-18 13:26
    Is there a way to use the Zener diode that would be more efficient? or as Phil said i should just use a LDO regulator.

    Since I don't have an education in electronics, it just seemed to me that dropping a couple of volts using a zener for a circuit that only takes 40mA would be what a zener is designed for, but obviously not. What is the overall easy to understand reason the zener voltage regulator is a bad idea in this circuit, because of the boost module? If for example you had a circuit with one 40 mA LED a resistor and a zener to regulate voltage, how is that different with reguard to the dc-dc boost? Assuming a zener voltage clamp would be a good circuit for varying voltage input to drive a 40ma LED.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Think Inside the box first and if that doesn't work..
    Re-arrange what's inside the box then...
    Think outside the BOX!
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2008-09-18 23:31
    I cannot make·a good recommendation without more information.
    ·
Sign In or Register to comment.