Update 19th August
There is a global supply problem with 1.5milliohm sense resistors. Neither Farnell nor RS Components have suitable devices in stock. Mouser has some but my latest order to them has been held up for some kind of customs reason. I don’t know why – never had a problem before. Potentially on track for receiving all components 25th August but it may slip by a day or two. I never realised how infuriating it is trying to ship lots of different items as a set of parts ! Please bear with me. Bernie GM4WZG
Bernard’s Overvoltage Board (named as such by my friendly local amateurs) is a module designed to sit between a 13.8V PSU and the equipment that it serves. It’s purpose is to react to sudden overvoltage situations by shutting down within microseconds and thus preventing subsequent damage to expensive equipment. It provides the same function as the old-fashioned crowbar circuit but in a much less brutal fashion.
To read my original article you will need to get your hands on the RADCOM journal. The best way to receive RADCOM is to ensure you are a member of the RSGB.
A blank PCB and components are available from the author on a first come first served basis. You will find them here. When they’ve gone, they’ve gone !
The original BOB was enthusiastically received by members of the local radio club and it does make an ideal club project. The module can be built into a PSU, or it can go inline in the DC cable to your amateur radio transceiver.
To construct it, surface mount skills are needed but there are not too many parts and given the guidance of a suitable mentor it might make an excellet project for those who want to learn those skills. There are two power MOSFETs on the board and they have solder pads underneath them which means that either a hot air rework station (available for £30 from Ebay), a hot plate (I’ve used a domestic £20 single plate one from Argos in the past) or oven are required for proper assembly. All other components can be soldered with care using a soldering iron.
For those that prefer, I have packaged all of the CAD files in a zip ready to be sent to a PCB fabrication house of your choice.
If you would rather that I send you a blank PCB then you can get them here. You’ll also want this project bill of materials:
C5 is a 0.22uF MLCC which is connected to the retry pin of the LTC4368. If instead this pin is directly grounded then the device will latch in the off state once the protection operates. Turn the power off and on again to restart.
Some datasheets here. .
The builder should decide on the desired Overvoltage Setpoint and select 1% tolerance resistors accordingly.
BOB is capable of more than Overvoltage protection. It can perform short circuit protection with a few caveats. Many amateur radio rigs cause a huge surge of inrush current when powered up. If you use the fast-acting short-circuit protection that BOB makes available to you then when using those rigs there might exist the unexpected consequence of shutting the power off at the moment of turning on the rig.
Fuses and the BOB Overcurrent Set-point. A 20 amp power supply is usually capable of supplying much more current than 20 amps for at least a several milliseconds if the load demands it. That is why circuit board traces can evaporate and semiconductors rip themselves apart when faults arise in load equipment. We typically protect a 20amp transceiver with a 25amp fuse. But what does that actually mean? Well a 25amp fuse, when subjected to 50 amps, will typically blow after anything between half a second and several seconds. See the table below. As many of us have found out to our cost, that can be quickly enough to protect the power supply and avoid a fire, but not quickly enough to protect the load i.e. our transceiver. BOB's primary purpose is to provide super-rapid overvoltage protection but the LTC4368 overcurrent (short-circuit) protection circuitry can be useful too - if we understand how to use it. Current that exceeds the set-point causes power shutdown. But the LTC4368 is fast - very fast. The set-point must take into account inrush current of the load equipment otherwise the very act of turning the equipment on might cause the LTC4368 to shut down. The inrush current comes mostly from the initial charging of the many capacitors in parallel found inside a typical HF rig. They may not be high value capacitors, but they are of low ESR and they are in parallel. The overall ESR can therefore be miniscule. Inrush currents can be very high, even if only for a few microseconds. The author's Icom and Kenwood rigs have inrush in the order of 25 amps for 0.5 millisecond so a 33 amp BOB setpoint is appropriate. The author has examined a Yaesu FT991 which exhibited inrush current of 72 amps for 0.5 milliseconds - see a full explanation and oscilloscope traces here. Yes, that is huge. Owners of FT991 and other similar rigs might need to set a particularly high setpoint or if they prefer, just omit the overcurrent protection altogether.
| Update 13/08/2021. After discussing this problem with another Yaesu user it occurred to me that there might be a better solution than just deciding not to use the built-in overcurrent protection. |
BOB will sense and close down in 8 microseconds when built according to the LTC datasheet. But by making the current sense part of the circuit into a kind of low pass filter it is possible to require a full 1 millisecond of overload before protection kicks in. This copes with the 70+ amps power on surge of the FT991 yet allows a 20 amp overload to be set. Any overcurrent more than 20 amps for more than 1 millseconds causes close down. Not as fast as 8 microseconds but a factor of 100-10,000 faster than any fuse. That has to be worth going for.
I’ve simulated it in Spice and it works a treat, exactly as designed. Not yet done a real life test – real life kind of gets in the way 🙂
I’ll document it fully here when I get the time and will also report any real live tests. The modification, by the way, is not easy to do on the existing circuit board. If I ever make another run then I can incorporate it properly.
The problem with traditional fuses. In a nutshell they are slow and unreliable compared t othe circuitry that they are supposed to protect. They often save the power supply but not the load.
|Fuse under test||Fuse Rating (A)||Time to Blow at stated rating||Time to blow at 200% rating||Time to blow at 400% rating|
|SinglFuse™ SF-3812F-T Series||25A||Possibly Never||60 Seconds||Not Stated|
|Multicomp Fastblow||25 A||Possibly Never||4 minutes||Not Stated|
|Cooper Bussman 500 Fast Acting||15A||Possibly Never||2 minutes||3 seconds|
|BEL SMD Fastblow||25A||Possibly Never||Not Stated||5 Seconds|
|LittelFuse ATOF® Blade Fuses||25A||Possibly Never||5 seconds||500 milliseconds|
Finally, let’s look at an interesting experiment that I have conducted several times to show how rapidly the overcurrent actually protection of the LTC4368 comes into play.
1. Take a good quality power supply capable of sourcing 20 amps and that contains a 25 amp fuse.
2. Connect it to a dead short.
3. Unsurprisingly the 25amp fuse blows (and we hope the power supply survives but don’t carry out this test yourself just in case!).
4. Replace the fuse (and the power supply if you blew it up!)
5. Now short it out with a forward biased 1n4148 signal diode.
6. Guess what? The diode blows, the fuse doesn’t. The diode “became” the fuse. That diode may well have been a component inside your rig – disaster!
7. Repeat the above tests but with an inline BOB having a set-point of 50 amps.
8. Connect a dead short.
9. Boringly, nothing happens other than BOB switches off the power.
10 Now try the experiment with the 1N4148 small signal diode. Short out the BOB with that.
11. Guess what? Boringly still nothing happens. BOB switches off the power before any damage occurs to the diode i.e. phenomenally quickly.
If the BOB setpoint is made substantially higher than 25 or 33 amps we provide both immunity to inrush current tripping yet we also make it unlikely that internal rig faults can propagate to cause trace burnouts and other irreparable damage.
I have had several questions and thought it might be useful to reproduce them here
How hot does BOB get when pulling 20 amps ?
The easist answer is to show you a thermal image after 5 minutes of 100W key down on the author’s TS890S which incidently, has rock solid thermal design.
At a temperature of 55 degrees this is just warm to the touch.
Which way around do the MOSFETs go ?
The MOSFET symbol on the PCB is partly obliterated so it is not obvious at first glance which way around the MOSFETS go. If you turning them upside down to see the MOSFETS from underneath and comparing that with the PCB footprint should make it obvous.
How do I solder the LTC4368 ?
Very careully ! It is small and pins easily bent. I use small amounts of solder paste on the PCB pads and a combination of hot air and a hot plate to make the joint properly. It is fiddly, but easily achieved with perseverence.
Is it possible to use Anderson Powerpoles ?
Yes. The spacing of the inner terminal block through holes is 7.62mm. The spacing of two Anderson Powerpole 15-45 amp PCB pins is 7.9mm. The difference is only 0.28mm which is not going to make any material difference when inserting and soldering. Alternatively put powerpoles on the end of short power leads soldered to the board.