Enginer PHEV Technical Information

A quick correction to your post
Output is 15.15A * 53.5V = 810.5W

 

I'm sure you mean 810.5W

Thanks for the update. Do you think there will be a noticeable difference between using 50ft 12G and a 25ft 12G cord? Would you recommend going to a 10G or a 15ft cord?

 

Is there any info yet on the DC/DC converter?

 

Could you take a picture of any key markings on the the Converter and charger. Model number, and the like. I can't find any docs on either, although the charger should be on thunder-sky or mottcell, but google might find it quicker than me.

11011011

 

OK... I looked at the picture and I thought I saw "Model" but it was really "Made in..."

Thanks for all the good pics.

 

Awesome information! I don't suppose there's a way of figuring out what is used to do the programming so we might tweak the output ourselves? Are the future enhancements something to interpret and display the data or is the unit currently outputting any information via that port that we might be able to interpret ourselves via some creative DIY?

 

Couldn't decide whether to post my question on the PHEV for dummies thread or this one, but since it's so Enginer specific, this might be the best bet. I pulled up my BTV gauge on my SGII and watch the battery voltage bounce around. Very informative. What I noticed is that putting at 240 V my battery does pull 12A, but only if it's at 7 bars already. If it's down around nominal 220V (3 bars), 240V will dump something like 50A. I would think that 50A draw on a the back end of a 240V DC/DC converter would come out to something like 12KW. More than enough to melt a 4KW DC/DC converter.

So I guess I assume the converter would have to be reading the nominal voltage on the line then step it up incrementally. Targeting 240V but stopping when it was pushing 12A. If only Converters were written in C it would look something like this:

Code:

maxV = 240;
maxW = 2880;
nv = calc_downstream_nominal_V();
start_v = nv++;
outputW = 0;
for(targetV = start_v; targetV < maxV && outputW < maxW; targetV++)
{
  set_downstream_V(targetV);
  sleepMicroSec(1000);
  outputW = calc_output_wattage();
}

while(true)
{
  outputW =  calc_output_wattage();
  if (outputW >= maxW) { targetV--; }
  else if (targetV < maxV) { targetV++; }
  set_downstream_V(targetV);
  sleepMicroSec(1000);
}

I just don't see how holding 240V consistently keep the converter within spec (ie, not melt it) when the OEM battery was down at 3 or even 4 bars.

Next question was back to regen. So if the OEM inverter sees the 240V (since the converter is driving 240V) and wants to do a regen it will step the voltage up to something like 245 or 250. At this point there is no longer a deltaV between the AUX pack and the OEM pack so the converter will effectively shut off. Once the converter comes off though, I would expect the voltage drop on the bus and a fairly substantial inrush until things even out. Kinda like a tug-of-war where one of the members from the other team (the converter) suddenly lets go of the rope.

Is this sound physics or is there something about current stabilization that eludes me? Do all these things magically equalize which is why this solution is so elegant?

11011011

 

This is completely a guess, but I would imagine that the converter functions more like a current limited voltage source. A lot of regulated voltage sources are really more like current sources than voltage sources anyway. A simple pass regulator has a current output device, and an op-amp that measures voltage and adjusts output current to keep the bus voltage where it should be in a feedback loop. A current limited voltage regulator would do exactly the same up to a certain preset maximum current. At that point the loop would "rail", maintaining a constant output current but allowing the output voltage to droop.

To put it in a more general way, a regulated voltage output is a non-linear device that can appear as if it has nearly 0 dynamic resistance over a certain range. That is to say that for a certain range of delta I, delta V is more or less 0. This approximates what an ideal voltage source is supposed to look like (which is why people use voltage regulators), but real ones like batteries almost never really do. Once you exceed a certain max I, the regulator can no longer supply enough current to hold V constant and you can be in danger of damaging something while it tries in vain to do so. A simple improvement in some regulators is to just shut themselves off when they sense I>Imax. A more sophisticated way is to add a "current limited" operating mode, as mentioned above. In this mode, once Imax is reached, the regulator behaves more like an ideal current source, outputting a fixed current regardless of voltage, so now delta I is zero for any reasonable delta V, and dynamic resistance is nearly infinite.

The above description is very similar to how many lead acid battery chargers work. When you first hook them up to a battery with a low resting voltage/SOC, they would pull a lot of current out of the charger if they could. For this reason many chargers start out in a current limited opertaing region, allowing the output voltage to sag to keep from trying to source too much current and damage the charger and/or battery. As voltage and SOC gets higher, the charger will transition into a constant voltage mode, where current will slowly taper to 0 as the battery reaches 100% SOC. In general voltage/current regulators, battery chargers, and dc:dc converters are all very similar devices.

Rob

 

BTW, what your c code basically describes is the feedback loop of a regulator. In hardware this is most often done in an analog fashion with an op-amp in more or less real time. The inputs to the op-amp are a known reference voltage, and the regulator output voltage. The op-amp's job in life is to adjust its output such that difference between the two inputs is zero. The op-amp output drives the current source device, and supplies exactly the amount of current needed to keep the output voltage the same as the reference voltage regardless of the load at the output. Since the circuit "loops" back around from the op-amp's output to its input this setup is referred to as a feedback loop, or closed-loop control.

Rob

 

You're right on here I believe. The filtering elements like capacitors keep the changes slow enough that the regulators can keep up, and keep the dV/dt gradual enough to avoid big dI/dt spikes that could damage something.

 

Balancing all 16 would certainly be preferable, but 8 is probably still pretty good. You can generally think of each battery as part of a random statistical distribution (like a bell curve). Most cells should be fairly close to the average. There will be some variation between individual cells, and in any group there will be a high and low outlier. The balancer I would assume evens this out the best it can within a group of 8. Assuming the distribution is fairly balanced (bell curve) like, a group of 8 cells averaged together should match another group of 8 cells averaged together much better than any two individual cells match. This should be particularly true if all the batteries in a set are from the same manufacturing lot, and have all been cycled identically together at all times. This is less true if you try to replace an individual weak/dead cell or upgrade a pack.

Does anyone know if the balancers are only active during charge, or do they also monitor cell voltages during discharge? Ideally a BMS should automatically disconnect the system when the weakest cell is depleted, or balance during discharge. The issue is that the batteries will all have slightly different capacities. Balancing on charge helps get all the batteries charged up to the same level, which is very important to prevent them drifting apart over time. However, the weaker cells in the pack will regularly be run to lower SOC than the stronger cells. Since lifespan is related to DOD, your cells which were already weaker to begin with are being aged at an accelerated rate.

One way to gauge how much of an issue this might be is to monitor the Voltage balance of all the cells in the pack when the pack is depleted, not just when its full.

Rob

 

Would you mind sharing the source of this information?

 

-

 

Re: Manuals

Quickly scanned the Battery Balancer manual for a temperature rating the only thing that was said was not to use the battery balancer in high ambient temperatures. I question what is the rated temp is. Generally, if no rating is given the rated value is low and the MFR is in CYA mode. The unit may only function at 115°F or below. I seen this before for other industrial components, AB for one.