X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Subject: Re: [geda-user] [off-topic] 24V automotive (truck) design issues
To: geda-user AT delorie DOT com
References: <eda2cdce-ccc4-67f4-ccde-6798043e5f41 AT linetec DOT nl>
 <B4FB33EE-C263-4E55-AFAB-63059CAC342B AT noqsi DOT com>
 <7c1717bb-1b7d-360f-d06a-71ee6983eea1 AT linetec DOT nl>
 <CAGqyy=bCkYxADbRhjfAeCkt2bOccMGUDFHMRy5myyXWjZ3s8_w AT mail DOT gmail DOT com>
 <e2fcdb04-5b05-351f-8c05-386c588b9ebb AT linetec DOT nl>
 <CAGqyy=aPKOLPG+=Wum8Vi=6f51hS+4rhjrz=UYedYkkPvUVU_A AT mail DOT gmail DOT com>
 <54e56571-e668-f382-5f18-7e8f579c8531 AT linetec DOT nl>
 <CAGqyy=ZnzCZm+=Yxe=Et0zqsPVHCG9TmyGtZnrFQfy8UiWknyQ AT mail DOT gmail DOT com>
 <30435625-5dca-7b6f-93af-a322faaea7a3 AT linetec DOT nl>
 <CAOuGh8_huApMbSoUOuTTDS10kn+ZiPOtoLfQCo3L8aiaDmFUPA AT mail DOT gmail DOT com>
 <CAGqyy=ZOCN0+Ko0qCeJhf_dHK72GF42UkCLUL8bEt3FMNKUHjg AT mail DOT gmail DOT com>
From: "Richard Rasker (rasker AT linetec DOT nl) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Message-ID: <e0051ed9-813f-a07e-f0c7-b2364422a33d@linetec.nl>
Date: Sun, 2 Dec 2018 15:52:44 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.2.1
MIME-Version: 1.0
In-Reply-To: <CAGqyy=ZOCN0+Ko0qCeJhf_dHK72GF42UkCLUL8bEt3FMNKUHjg@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Hi Luis,

Op 29-11-18 om 16:18 schreef Luis de Arquer (ldearquer AT gmail DOT com) [via 
geda-user AT delorie DOT com]:
> Richard,
>
> Your proposed circuit seems OK to me. Just make sure the feedback is
> strong enough so the MOSFET never stays in active zone (especially
> since it has low-ish gate threshold voltage).

Well, it turns out that the circuit was spot-on with the values given: 
disconnect at 37 volts, and reconnect at 34 volts.

> We considered using this from Texas Instruments in some project -we
> didn't need to eventually:
> http://www.ti.com/lit/an/snva190b/snva190b.pdf
>
> Maybe your problem was, after all, with the negative 1ms - 2ms pulse.
> That one is pretty bad, and TVS energy spec is not too far from it
> (600W 1 ms 25deg IIRC, worse with higher temperature).

Yes, that could very well have been the cause. Unfortunately, someone at 
the truck company messed up last Friday, as they had the wrong truck for 
me to examine -- this one had an almost perfect onboard supply that 
showed no peak voltages at all, no matter what we tried, and my PCB's 
were just fine on this one.

The truck that blew up my boards was somewhere in Denmark, some 500 
miles away, so I'll have to arrange another measurement session with 
that one.

> Also, since your current consumption is not too high, maybe you could
> consider adding some line resistance, if cost is an issue for you.
>
> Keep us posted :)

After deliberations with the customer, we decided that I create a 
separate supply unit, designed to handle everything outlined in the ISO 
documents.

The main reason was that I didn't have room on the LED PCB's for large 
components (3 mm maximum height), precluding the use of largish 
capacitors for uninterrupted operation during longer voltage peaks. And 
because they also wanted to power multiple LED boards, I chose the 
TPS54260Q buck regulator, protected by the disconnect circuit (with a 
far heavier SQJ431EJ PMOS device).

So alas, I still don't know the exact reason why the first boards blew 
up, other than that it must have been a voltage peak (> 40 volts) with 
sufficient I*t to blast the TVS diode. If and when I do find out, I'll 
be sure to let you know.

Anyway, thanks again for your interest in and input on this matter!

Best regards,

Richard Rasker