Prev: Mitashi 4 GB Mp4 Player - MW156
Next: aluminum heatsink anodizing
From: John Devereux on 19 Dec 2008 13:13
Jim Thompson <To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> writes:

> On Fri, 19 Dec 2008 14:48:21 +0000, nospam <nospam(a)please.invalid>
> wrote:
>
>>Raveninghorde <raveninghorde(a)invalid> wrote:
>>
>>>For the benefit of the OP.
>>
>>>This circuit is not suitable for your application as drawn.
>>
>>The first question is is it acceptable to shut down during a load dump.
>
> Depends on the circuit configuration. Many of my ignition system
> designs protected themselves by turning off appropriate devices
> _during_ the load dump pulse.
>
>>If
>>it is you can buy LDO regulators with reverse battery and 60v load dump
>>protection built in.
>>
>>I presume 60v is enough for 12v automotive systems else they wouldn't
>>bother trying to sell them?
>
> I haven't looked "under the hood" with a 'scope in years, so I'm not
> familiar with what load-dump controls that might have been added.
>
> But I can assure (scare) you that I ran storage 'scope investigations
> in the '60's that showed load dump pulses as high as 400V :-(
>
> And I have chips turned to beads of silicon to prove it ;-)
> (Inductance-driven arc discharge across the chip.)


I was thinking of making a load dump simulator/generator "one day".

There is a circuit in the datasheet for the "LDP24M" (p4).

<http://www.digchip.com/datasheets/download_datasheet.php?id=502684&part-number=LDP24>

The energy and currents involved are huge!

I have seen voltages specified up to ~150V.

47,000uF charged to 150V dumped through 2 ohms is a pretty serious
pulse. I've actually gone as far as snapping up a cheap box of 200V,
1500uF electrolytics to build the capacitor bank required.

--

John Devereux
From: nospam on 19 Dec 2008 14:51
Raveninghorde <raveninghorde(a)invalid> wrote:

>MIL-STD-1275 for 24V vehicles specifies a single fault spike of 250V
>for 70us with 100V surges of 50ms. So 251V is allowed to fry the
>electronics.

The 100v load dump requirement is intended to cover cases where the vehicle
is operated with no battery connected.

Higher voltages are low energy spikes which can be absorbed, the load dump
tests can source about 200J. Your typical 1500W transient suppressor can
absorb about 1.5J.

We were talking about cars on 12v not military vehicles on 24v.

If 100v is good enough for 24v military systems with no battery connected a
60v requirement on 12v systems seems quite adequate.

--
From: Jim Thompson on 19 Dec 2008 16:35
On Fri, 19 Dec 2008 18:13:09 +0000, John Devereux
<john(a)devereux.me.uk> wrote:

>Jim Thompson <To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> writes:
>
>> On Fri, 19 Dec 2008 14:48:21 +0000, nospam <nospam(a)please.invalid>
>> wrote:
>>
>>>Raveninghorde <raveninghorde(a)invalid> wrote:
>>>
>>>>For the benefit of the OP.
>>>
>>>>This circuit is not suitable for your application as drawn.
>>>
>>>The first question is is it acceptable to shut down during a load dump.
>>
>> Depends on the circuit configuration. Many of my ignition system
>> designs protected themselves by turning off appropriate devices
>> _during_ the load dump pulse.
>>
>>>If
>>>it is you can buy LDO regulators with reverse battery and 60v load dump
>>>protection built in.
>>>
>>>I presume 60v is enough for 12v automotive systems else they wouldn't
>>>bother trying to sell them?
>>
>> I haven't looked "under the hood" with a 'scope in years, so I'm not
>> familiar with what load-dump controls that might have been added.
>>
>> But I can assure (scare) you that I ran storage 'scope investigations
>> in the '60's that showed load dump pulses as high as 400V :-(
>>
>> And I have chips turned to beads of silicon to prove it ;-)
>> (Inductance-driven arc discharge across the chip.)
>
>
>I was thinking of making a load dump simulator/generator "one day".
>
>There is a circuit in the datasheet for the "LDP24M" (p4).
>
><http://www.digchip.com/datasheets/download_datasheet.php?id=502684&part-number=LDP24>
>
>The energy and currents involved are huge!
>
>I have seen voltages specified up to ~150V.
>
>47,000uF charged to 150V dumped through 2 ohms is a pretty serious
>pulse. I've actually gone as far as snapping up a cheap box of 200V,
>1500uF electrolytics to build the capacitor bank required.

A/C thermostat driving compressor clutch will do it ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food
From: John Devereux on 19 Dec 2008 16:53
Jim Thompson <To-Email-Use-The-Envelope-Icon(a)My-Web-Site.com> writes:

> On Fri, 19 Dec 2008 18:13:09 +0000, John Devereux
> <john(a)devereux.me.uk> wrote:
[...]
>>
>>I was thinking of making a load dump simulator/generator "one day".
>>
>>There is a circuit in the datasheet for the "LDP24M" (p4).
>>
>><http://www.digchip.com/datasheets/download_datasheet.php?id=502684&part-number=LDP24>
>>
>>The energy and currents involved are huge!
>>
>>I have seen voltages specified up to ~150V.
>>
>>47,000uF charged to 150V dumped through 2 ohms is a pretty serious
>>pulse. I've actually gone as far as snapping up a cheap box of 200V,
>>1500uF electrolytics to build the capacitor bank required.
>
> A/C thermostat driving compressor clutch will do it ;-)

I guess I am a bit too electronics orientated - because I have no idea
what those words mean (when put in that order)!

--

John Devereux
From: Jasen Betts on 20 Dec 2008 04:45
On 2008-12-19, IanM <Invalid(a)totally.invalid> wrote:

>
> So replace M1 with a diode as its only there for the reverse protection
> and the OP has plenty of headroom with a 7V regulated output even if one
> needs to run right down to 10V in.

> I'm more concerned about M2. Given a suitable FET, I suppose it
> *should* survive provided the trigger point for the TL431 is less than
> its Vgs rating and the TL431 protect itself as it triggers but if you
> aren't muntzing it, a Zener connected gate-source would prevent
> transient stress on the gate oxide. If you *are* muntzing it, make M2
> the pass transistor in the regulator as well.

also you need to increase R3 to about 3K to save the TL431 from having
to conduct 1A during faults.

to compensate you may need to add some voltage drop (three 1m914's in series)
in series with R4 (which should also be increased to about 3K)

First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4
Prev: Mitashi 4 GB Mp4 Player - MW156
Next: aluminum heatsink anodizing