Today's Message Index:
----------------------
1. 07:32 PM - Re: OVM-14 MkIII development (update) (Eric Page)
Message 1
| INDEX | Back to Main INDEX |
| NEXT | Skip to NEXT Message |
| LIST | Reply to LIST Regarding this Message |
| SENDER | Reply to SENDER Regarding this Message |
|
| Subject: | Re: OVM-14 MkIII development (update) |
There haven't been any updates on Bob's OVM-14 Mk III overvoltage protection module
since Feb 2024 so I took a look at it to see what I could do. I found some
minor things that might have been problems, so I made a few small changes:
1. Looking at the datasheet for the crowbar SCR... [https://tinyurl.com/yjvrdtfx
(https://tinyurl.com/yjvrdtfx)] ...the specification for gate triggering current
states that it could take as much as 15mA to trigger. As configured in the
last schematic that Bob posted... [https://tinyurl.com/3jeddsky (https://tinyurl.com/3jeddsky)]
...with the gate drive resistor (R14) of 2.49k ohms connected
to the bus, 15mA gate current would not be achieved until the overvoltage
exceeded 37V.
CHANGE: R14 reduced to 620 ohms and connected to the 10V supply rail instead of
the bus. This provides 16mA to trigger the SCR's gate, while preventing an overcurrent
through the output of comparator U11.B (maximum 20mA).
2. At a bus voltage of 14V, current-limiting resistor R1 (392 ohms) would dissipate
of about 1/3 watt, and would run hot.
CHANGE: R1 increased to 470 ohms and pass transistor Q17 added. This provides
a sturdy 10V supply rail for the circuit, including adequate gate current to trigger
the SCR. Continuous power dissipation in R1 will be reduced to 0.03W.
At the high end of its 1% tolerance (475 ohms), with bus voltage at 12V, R1 will
provide 3mA to U3 and the base of Q17, more than double the minimum required
for regulation. At the low end of its tolerance (465 ohms), R1 will protect
U3 against exceeding its absolute maximum current rating up to a ridiculous
bus voltage of 81V.
3. The trip delay capacitor (C12) is a ceramic type, which may not be stable in
a timing application due to the effects of DC bias, temperature and aging.
CHANGE: C12 changed to a 5% tolerance solid tantalum type.
-----
With that done, I created a board layout that's very similar to Bob's and is the
same size (0.7" x 2.2"). It still uses all through-hole parts except the SCR,
which is large for a surface mount component and is relatively easy to hand
solder. I moved the SCR to the top side of the board and added a 2-position
0.25" quick-connect terminal to make installation easy (wires can still be soldered
directly to the board if preferred). I also changed to a "piano key" style
DIP switch to activate the maintenance mode, which makes the switch accessible
at the end of the assembly after heat shrink is applied.
After building a prototype, I found that the resistors that set the trip voltages
didn't quite work as calculated, so I tweaked them experimentally to find the
values that provide a normal trip at 16.0V (R8 = 2.61k) and a maintenance test
mode trip at 13.7V (R10 = 69.8k). To ensure that maintenance test mode works
correctly with all battery types, I bumped the test mode trip point up by
0.2V, from 13.5V to 13.7V. This will accommodate the higher resting voltage of
a fully charged LiFePO4 battery (typically 3.4V per cell, or 13.6V for a 4-cell
battery).
Finally, I assembled four additional prototypes and tested all five. In each case,
the modules tripped at the desired voltage and in the desired time, blowing
a 5 amp fuse with no discernable heating of the SCR.
Full documentation for the project is attached to this post. You'll find links
on page 7 to purchase circuit boards and components.
Read this topic online here:
http://forums.matronics.com/viewtopic.php?p=515348#515348
Attachments:
http://forums.matronics.com//files/nuckolls_ovm_14_mkiii_rev_1d_636.pdf
Other Matronics Email List Services
These Email List Services are sponsored solely by Matronics and through the generous Contributions of its members.
-- Please support this service by making your Contribution today! --
|