[Tesseract]

Originally Posted by DIYguy

Sounds like it would work ok... only thing is, using my exaple, I would have to get the soldering iron out each time I wanted to change the valet setting...which admittedly would likely not be often.

Or you could replace the fixed resistor with another pot (wired as a variable resistor) so you could tinker with the value until you are happy.

Originally Posted by DIYguy

I suppose the cream would be a whole new set of operating perameters... ramp rate, max amp, min battery ... although one likely doesn't need all of that.

Changing the ramp rate would useful, but only if it's much above 500A/s in the first place, so we might hard code that into valet mode just to keep the home page as uncluttered as possible.

Originally Posted by DIYguy

(i could always throw my other controller in each time ... ouch!)

No comment!

[DIYguy]

Originally Posted by Tesseract

Or you could replace the fixed resistor with another pot (wired as a variable resistor) so you could tinker with the value until you are happy.

Ya, good idea. That would work.

[bipole]

Originally Posted by Tesseract

QC/calibration for each controller is done at 252V/1000A, but the design was proved at 300V and 1000A....

I must apologize, I quoted the other controllers' KW without reporting the actual specs.

Z2K max battery current:200V: 1900 A= 380 KW
250V: 1835 A = 458 KW (interpolated)
300V: 1770 A = 531 KW
400V: 1600 A = 640 KW but the battery pack has to sag to get any amps out, so this is virtually unobtainable.

So, even the 300V spec would take an extremely stiff battery pack to hit those numbers with the specified 348V max nominal input.

I just thought it would be prudent to eat a little crow and report the actual specs.

[Bowser330]

Originally Posted by bipole

I must apologize, I quoted the other controllers' KW without reporting the actual specs.

Z2K max battery current:200V: 1900 A= 380 KW
250V: 1835 A = 458 KW (interpolated)
300V: 1770 A = 531 KW
400V: 1600 A = 640 KW but the battery pack has to sag to get any amps out, so this is virtually unobtainable.

So, even the 300V spec would take an extremely stiff battery pack to hit those numbers with the specified 348V max nominal input.

I just thought it would be prudent to eat a little crow and report the actual specs.

interesting side point...the EHV models of Zilla state 348V max however the S10 Smokescreen was using one with a pack voltage of almost 400V...

maybe some type of custom setup...

[Tesseract]

Originally Posted by Bowser330

interesting side point...the EHV models of Zilla state 348V max however the S10 Smokescreen was using one with a pack voltage of almost 400V...

maybe some type of custom setup...

Nah, Bowser, both the Zilla and Soliton1 use 600V IGBTs, which are intended to be used in products that run directly off of rectified and filtered 240VAC, which gives a nominal DC bus voltage of 340V, so 400V ain't exactly "instant death" territory. If your power source is a battery pack - i.e. no worries about lighting, load dump transients, etc. - you can bump the voltage up to as high as 480V (80% of max). This assumes that the huge increase in EMI from operating at a higher voltage doesn't wreak havoc, of course...

(A side note worth mentioning: the legal limit for terrestrial broadcast radio stations is 100kW and no one would expect an electronic device to function fine right next to such a transmitter's antenna, but that's what we expect of the control boards inside a Soliton1, Zilla, etc... )

Unfortunately, increasing the voltage also cuts into the current rating because it greatly increases the switching losses - perhaps to such an extent that there is no net increase in actual power output! Long term reliability is also decreased regardless of how well spikes and such are managed. Modern IGBTs and MOSFETs have a "square" Safe Operating Area (SOA) curve, but only if you keep the actual die/dice at 25C, which is generally unrealistic unless the ambient temp is -20C or colder! So, as you increase operating voltage there comes a point at which you have to start decreasing operating current OR duty cycle/pulse width, and these are not widths measured in seconds, folks, but in milliseconds or even microseconds (e.g. - the IGBT modules used in the Soliton1 can withstand 3000A for 6us, which is nice for surviving a short circuit for just long enough to shut them down safely, but not at all a useful current rating!) See the pic illustrating a typical SOA curve, which applies to the controller itself as well as the devices that do the actual switching.

So as is always the case, everything comes down to finding the point of maximum power for the batteries, controller and motor. Sometimes it makes sense to go with a higher pack voltage composed of smaller (lower Ah) batteries and let it sag down to a voltage the controller is happier with; sometimes better results are obtained with a lower voltage but stiffer pack. If there was a single answer to this problem then drag racing wouldn't be any fun, now would it?

[DIYguy]

Tesseract....
I am preparing for my Soliton1 which should be here Monday or Tuesday. I have a couple of questions....

Regarding the programmable outputs, I read they could be used to drive meters..... but couldn't find much info on this. Are there internal shunts for driving current meters? (likely not..) What can I program to these outputs? I want to have both battery and motor amp meters.... at least in the beginning. I have two shunts that I can wire up... but wasn't sure what exists currently.. (no pun intended) lol

Secondly, regarding the water cooling.... Is there any recommendation on on cooling capacity? I read the 1 - 2 gpm... but obviously the method of cooling the fluid can change a lot.... ie; reservoir only vs small rad/fan combination. Perhaps Dimitri has some feedback on this? (I did read your recommendation on ambient temp considerations)

Thanks,
Gary

[EVComponents]

Originally Posted by Bowser330

interesting side point...the EHV models of Zilla state 348V max however the S10 Smokescreen was using one with a pack voltage of almost 400V...

maybe some type of custom setup...

The nominal voltage of an EHV is 348V.
The max voltage we recommend is 375V.

They are tested higher than that though. There is a software cutoff that should protect the controller.

[Qer]

Ok, meters. This is a work in progress (very much right now) but in version 1.0 it looked like this:



This is because I screwed up and there's a buffer in the web server that gets truncated, so lets just ignore that version (*ahem*) and look at beta 373 instead, which is what we deliver atm (ideally I'm against releasing beta versions to the public, but when the beta is better than the original version I can live with making an exception ):



This got typically too complicated and messy, not to mention too confusing so we've scaled it down a bit. The idea was that if you're going to use an ordinary panel meter you can use the 0-10 Volt range and just add a resistor in series to convert the voltage to current but apparently there's car meters (like tank meters) that use 0-100% of the 12 Volt for the full range and then there's those that work the opposite way (100-0%) but, well, it got so messy and confusing that we've decided to simply scrap all these choises and only keep the 0-100% and leave it to the DIY'er to convert the signal to something useful (and if you run a DC/DC, which we recommend anyway, the voltage will vary so little that you can still use the resistor approach).

So right now (in rc379) it looks like this instead:



Much cleaner. What you can't see here (because it's not implemented yet, work in progress as I said) is that some of these parameters will activate other input fields in the web interface where you can enter the preferred parameters and others will use already existing settings. For example, motor current will use the maximum motor current you've entered for the motor so if your max motor current is 800 Ampere the scale will be 0-800 Ampere for 0-12 (or 0-13.5 or whatever) Volt out and the same goes for battery current of course.

Power is one I haven't really decided on yet, but one way or another it will use some kind of setting to determine the scale since it will depend a lot on your setup what range you'd like to see. A commuter might like 0-60 kW but someone with a sports car probably want a wider scale, otherwise it might be smack at 100% as soon as you hit the throttle, not very useful.

Controller temperature is always 0-100% for 0-100C. Typically it should stay on the lower half of the scale...

DoD-level is a new one which definitely is work in progress, right now it only outputs zero since I'm working on the code. DoD-level is probably the wrong name since the plan is that it should work as a fuel tank meter, ie 100% is good, 0% is bad. "Charge level" perhaps? Name suggestions are welcome.

RPM is the only output that doesn't output a voltage since it's meant to be used with an existing RPM-meter. Right now it only outputs 2 ppt (ie targetted for tachometers from 4 cylinder engines) but the plan is that it should handle 2, 3 and 4 ppt. One question I've always wondered is what a tachometer for a 5 cylinder engine uses? 2.5 ppt?

Also; all outputs (except RPM) are PWM outputs. If you connect them to an ordinary meter it won't matter since it will average out nicely by itself (like your motor averages out the PWM), but if you want to use digital gauges this might mean that you'll have to use a RC-filter to get a stable reading.

Did this cover your questions? That is except cooling. I'll leave that one to Tesseract. It's physics.

[DIYguy]

Wow...well, yes. At this point I guess knowing that there is/will be meter outputs.. I don't have to wire the big shunts in the circuit. That's all I need at this moment. I'm more of an mechanical guy, so interfacing the meters later will likely drive the need for some assistance.

[Qer]

Originally Posted by DIYguy

Wow...well, yes. At this point I guess knowing that there is/will be meter outputs.. I don't have to wire the big shunts in the circuit. That's all I need at this moment. I'm more of an mechanical guy, so interfacing the meters later will likely drive the need for some assistance.

Nope, no need to measure the currents. Again...