nicknorman

I’ve asked on the other forum of someone who moors at Colehurst. Apparently the farmer is very old now and not been seen around recently. The previous phone number no longer seems to work. His daughter is perhaps taking the reins so he will try to find out her phone number.

Well it can be there. If you knew anything about control theory you would know it is just a matter of having the right parameters / coefficients in the feedback loop, and some non-linearity - ie quick to reduce field current (when you chop the throttle to idle) and slow to increase the field current when the rpm increases, which is easy to do in software. Mine did hunt very slightly when I first commissioned it, but I tweaked the parameters and now it doesn’t.

Well first we need to know exactly what it is for. Is it just to keep starter and BT battery topped up during long periods of non-use? Or is it also to charge the BT battery after it has been used?

CANBUS tends to use different bit rates from the RS232 “doubling” series. I would expect 125Kb/s, 250Kb/s, 500kB/s, 1Mb/s. But it could be anything. Do you have an oscilloscope? In which case you should be able to see the bit rates. You don’t need a spectrum analyser.

It’s not intended for battery charging because it has a fixed output voltage - no absorption - float change of voltage. But for the application you intend I think it will be ok, just set it to a float voltage such as 13.5v. It will be on constantly and takes a minimum of 50mA (1.2 Ah per day) but presumably the intention is that the solar will easily cope with that.

If coolant hasn’t been changed for 10 years there might be some gas in the system (product of corrosion) and I know from experience it doesn’t take much of an air bubble to trip the overheat sensors. I’d check for any air in the radiators etc. What happened with ours was I suddenly noticed the header tank much fuller than it should have been, transpired there was quite a lot of “air” in the system which had displaced the water. Yes I had left it too long before changing the coolant! Since I did that, no repeat of the problem.

Low coolant level creating an airlock that causes overheat shutdown?

Isolated is fine. Non-isolated is cheaper! The Dc dc charger will need negative connections. Just one for the non-isolated versions but to both battery negatives for the isolated version. Isolated vs non isolated is about differences between the voltages of the negatives. The positives will be at slightly different voltages but that is not relevant.

Yes. Well not exclusively by river, but the Erewash canal which joins on to the river, takes wide beam boats and the boatyard is at the far end. Quite a few (manual) locks but it’s pleasant enough.

Not really. The voltage for gel is fixed at 14.2v bulk and 13.8v float, the latter being rather on the high side. But the main problem is the temperature compensation, which as far as I can tell can’t be disabled. So at 5C, the voltages become 14.8v and 14.4v which is much too high for Li. Of course this doesn’t matter until the batteries get fully charged. If you always turn off the charger before the voltage gets too high, then fine. But humans make mistakes. Also true that the Fogstar BMS will turn off charging, but at low temperatures the float voltage will keep the Li at the top of the knee, and this isn’t good for Li longevity. No doubt it would last a good few years but the answer to your question “is this charger compatible with Li” is no.

No you don’t need the isolated version. The isolated version is for where the battery negatives are at different voltages, eg if you wanted to charger the upper half of a 24v battery with a 12v B2B. In your case all the battery negatives are at the same voltage (shunts ignored) and so non-isolated is fine. On the subject of shunts, make sure you connect it to the non-battery side of any leisure battery shunts. Edit: just to put it more clearly, the non-isolated version has input and output negatives connected together. The isolated version has the output floating, ie the Input and output negatives are not connected together and can be at different voltages.

Well since Ian is sulking and no longer playing, I might as well finish off the technical aspects of duty cycle vs output voltage. It’s all very fascinating! It is correct that at medium - high loads, the duty cycle can remain fixed as the load changes, with the same output voltage (efficiency losses disregarded and assuming fixed input voltage). This is known as “continuous” mode, since the inductor is continuously passing current. However as the load decreases, inductor current on each cycle hits zero - known as discontinuous mode. In this mode, duty cycle has to be adjusted down to maintain the desired output voltage. Which is why I said earlier that his statement "Without any control loop feedback you can draw as much DC current as you want at the output and the voltage won't change (ignoring small series resistances), because the supply is a voltage mode converter, not current mode." was wrong, as anyone familiar with SMPS design would know. Dealing with discontinuous mode and duty cycle limits is a key issue in SMPS design and this is why I can’t see how he could possibly have ever been involved in “ground up” design of SMPS. And whilst it is true that in continuous mode, Vout /Vin =D, it is also true that Iin/Iout = D. So this equation is no grounds for claiming that an SMPS is a voltage mode converter, since it could just as equally be called a “current mode converter”.

https://www.ajfenders.com/ I think would make them bespoke if what they had in stock wasn't to your liking.

OK well full disclosure, I dislike Arduinos and haven’t done anything with USB. USB is a bit complicated to DIY from scratch. However it seems to be just a case of downloading the relevant library, MIDIUSB, and using that to send the midi command bytes that you have built up. There is some description of how to use MIDIUSB here https://github.com/arduino-libraries/MIDIUSB/blob/master/examples/MIDIUSB_write/MIDIUSB_write.ino But I think you would have to use a proper IDE to programme the Arduino, not Tinkercad. Edit: although it seems you can add non-Tinkercad libraries "in line", see here: https://www.tinkercad.com/help/circuits/include-an-arduino-library

However I am certain Ian can. In fact he designed them.

Yea ok I get it, I will stop boring the rest of you.

You would need to build a MIDI channel voice message for each action, and send it via a UART at the correct baud rate (31,250bps, which is a bit “non-standard”). A voice channel message consists of 3 bytes of data which include the status byte which has the channel number and the action (note on, note off etc), which key (ie which button), and the velocity (but you are probably not going for velocity detection). So for each button pressed you would populate a 3 byte string with the fixed data of channel, note on and velocity for the first and third bytes, and insert the appropriate note (ie which button) in the second byte, then send it via the UART. For each button released you would do likewise but with note-off. See here for example of midi voice channel message format. https://cmtext.indiana.edu/MIDI/chapter3_channel_voice_messages.php

I had a look at the AD tech note, it’s not very well written so I can see why, if you were hastily looking at the internet to see how SMPS worked, you got confused. Here is the buck converter diagram: You say that Vout = Vin x Duty cycle. Your confusion arises because you missed the bit in the tech note that said “under steady state conditions”. Consider the case with no or insignificant load, a 12v supply and a 50% duty cycle. You say the output will be 6v. But even the briefest cursory examination shows that each time the MOSFET turns on, there is some voltage across the inductor which will push some charge into the capacitor thus raising its voltage. When the MOSFET is turned off a bit more charge flows into the capacitor via the inductor and diode, and its voltage increases further. With no load the output capacitor has no way to discharge and the capacitor charging process continues until its voltage matches the input voltage whereupon when the mosfet is on, there is no voltage difference across the inductor. So there is no way that a 50% duty cycle, or even a 10% duty cycle, can give an output of 6v under minimal load conditions. It would be 12v or very close. The duty cycle has to fall to zero, or virtually so, to maintain 6v under no or minimal load. Every vaguely competent electronic engineer knows this. As to the rest of your story, I don’t believe it.

Well I now think that it isn't true. You couldn't be even a slightly competent electronic design engineer and make the statement he just did, and he repeated it.

Let's hope not, because Ian doesn't have the first idea how an SMPS works. And clearly not, because he would have been sacked on day 1. But I'm guessing you want to believe him over me. Which is fine. Lots of people believe things that aren't true, such as the existence of their imaginary friend (aka their god). It's mostly harmless. Well, except for believing in one's imaginary friend. Which is the cause of most wars, death and suffering.

Yes but they are designed by large corporations employing specialists in the field.

I have often wondered about the fact that you have said you have designed just about anything that ever gets mentioned, covering a wide range of disciplines from large scale IC design, rf design, various other specialised things, and now SMPS design. Most people using SMPS would buy a chip and equip it with the various components specified on the data sheet - but no you apparently have designed it from first principles (for some reason). And yet it is now clear that you don't understand the very basics of SMPS. You say "Without any control loop feedback you can draw as much DC current as you want at the output and the voltage won't change (ignoring small series resistances), because the supply is a voltage mode converter, not current mode." Google AI says: Yes, the duty cycle of a Switch Mode Power Supply (SMPS) does vary with the load, as the internal feedback loop adjusts the duty cycle to maintain a constant output voltage despite changes in current draw (load) on the power supply; meaning, when the load increases, the duty cycle will typically also increase to compensate and deliver the required power. Key points about SMPS duty cycle and load: Feedback mechanism: A feedback circuit within the SMPS continuously monitors the output voltage and adjusts the duty cycle of the switching transistors to maintain a stable output voltage regardless of load fluctuations. Duty cycle definition: Duty cycle refers to the ratio of "on time" to the total switching period, essentially how long the switching transistor is turned on compared to the time it is off. Higher load, higher duty cycle: When the load on the power supply increases, the feedback circuit will respond by increasing the duty cycle to deliver more power. All of which concurs with my view. This might be against the forum rules, but I think it is necessary to expose you. It seems to me that you are a fraud.

On an iPhone there is a setting for wifi calling on or off. But even with it on, I don't think it will give you free calls because you would still be using the phone's network provider to route the call, and you are using the phone number they provided. Certainly Google seems to imply that most phone service providers will charge you for WiFi calls. And of course all you are doing is eliminating the first part of the call routing - from your phone into their system. Their system still has to route the call over their or another provider's network or the landline system, and that is not free.

I’m not sure it is possible, but hopefully someone will prove me wrong. Of course one option would be to put the router SIM into the phone to make the call, but that would be a PITA.

Most smartphones have Wi-Fi calling, however as far as I can tell the phone network provider typically charges the call as if it had been made over their cellular network. A free options would be WhatsApp calling, but the destination has to be on WhatsApp too. Same for Skype.