relay test

[John Larkin]

https://www.dropbox.com/sh/2dt16zi77k4ljos/AAA9jNP8c5cX3NlnZG2u4uQga?dl=0

This seems pretty fast for a DPDT 8 amp 250 VAC relay.


I think this is public. Dropbox makes a positive effort to be
confusing.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[Klaus Kragelund]

We use a similar relay

It has 8ms turn in time, and we have tested for 1million operations. One of the guys wrote a program for an ESP32 so we can monitor it real time

Why are you using a TVS? To have it open faster? (we just use a tiny diode, current is maximum the same as relay peak current

Cheers

Klaus

 

[John Larkin]

On Tue, 25 Feb 2020 11:49:22 -0800 (PST), Klaus Kragelund
<klauskvik@hotmail.com> wrote:

We use a similar relay

It has 8ms turn in time, and we have tested for 1million operations. One of the guys wrote a program for an ESP32 so we can monitor it real time

Why are you using a TVS? To have it open faster? (we just use a tiny diode, current is maximum the same as relay peak current

Cheers

Klaus

The actual driver will be a TI TPIC6595, which has a flyback clamp
voltage around 50 volts. The TVS is a pretty good model. TVS vs no TVS
(mosfet avalanche?) doesn't seem to affect contact timing much. A
clamp diode across the coil really does.

It makes a lot less audible noise with a diode across the coil.

My customer is asking if we can simulate a transient open circuit
fault, and it looks like we can offer him a programmable open-circuit
of maybe 10 milliseconds and up. We can do that in software, in a 1 ms
IRQ kicking the SPI into the string of TI chips. We'd fudge the
customer time request a bit, subtract 5 or 6 ms maybe.



--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[Lasse Langwadt Christense]

tirsdag den 25. februar 2020 kl. 20.49.26 UTC+1 skrev Klaus Kragelund:

We use a similar relay

It has 8ms turn in time, and we have tested for 1million operations. One of the guys wrote a program for an ESP32 so we can monitor it real time

Why are you using a TVS? To have it open faster? (we just use a tiny diode, current is maximum the same as relay peak current

"discharging" the coil with 48V-12V vs. the ~0.7V with a diode is faster

with right FET it can do the clamping, most drivers for things like fuel injectors are done like that to get fast closure

 

[John Larkin]

On Tue, 25 Feb 2020 12:58:28 -0800 (PST), Lasse Langwadt Christensen
<langwadt@fonz.dk> wrote:

tirsdag den 25. februar 2020 kl. 20.49.26 UTC+1 skrev Klaus Kragelund:
We use a similar relay

It has 8ms turn in time, and we have tested for 1million operations. One of the guys wrote a program for an ESP32 so we can monitor it real time

Why are you using a TVS? To have it open faster? (we just use a tiny diode, current is maximum the same as relay peak current


"discharging" the coil with 48V-12V vs. the ~0.7V with a diode is faster

with right FET it can do the clamping, most drivers for things like fuel injectors are done like that to get fast closure

The TI driver has, essentially, a 45 volt zener drain to gate on each
of its output fets. We'll just be careful to keep the relay driver
traces away from any logic signals, although they are probably fairly
slow.

My test board doesn't model the TI chip for slew rate.

I have just let the mosfet avalanche in other applications. Some are
rated to do that safely. SiC can usually be allowed to avalanche too;
GaN usually not.




--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com