DIY Electric Car Forums banner

dc-dc converter

28K views 141 replies 33 participants last post by  favguy 
#1 ·
I just found some dc-dc converters on ebay, I'm looking at a Vicor V1-251-09 200v to 12v model and am wondering if anyone is familiar with this one. I sent a request to the seller for the datasheet but while I'm waiting I figured I would post it here .
 
#41 · (Edited)
Electricar,

Good point about the diodes, they could be the weak link, will have to keep an eye on that one.

Cruisin,

Having tested one, I can confirm laptop power supplies are isolated. Although I absolutely agree that safety must not be compromised to save money, If properly fused on the input and output, I see little risk to either the pack or the car in trying this. As for what you are saying about items proved to work and sold by EV suppliers, this isn't always true. Take the vacuum pump kits as an example, they are very expensive and although they work, make an awful racket in use, The OEM VW/Audi/Landrover etc. vacuum pumps are far cheaper and run almost silently, I presume you prefer the former?!

This is a technical discussion forum to discuss matters such as this topic and help one another, I think your comment about getting attention is inappropriate. I come here for help, if you can give me a valid technical reason why this is a bad idea, I'll happily throw it out! Also, if this is "off the wall" may I remind you what we are all doing here!! Most of the population probably think everything we discuss here is "off the wall", by your logic, we may as well all give up and go and buy a Leaf!

DJBecker,

Thanks for the input, can you elaborate on the possible effect of paralleling these, regarding power output, would diodes on the +ve of all units help, although I suspect they already have one internally?

For what they cost, I might just buy four of them, parallel them up and apply 10A/20A/30A loads and soak test them at each for a few hours.
 
#42 ·
Electricar,

Good point about the diodes, they could be the weak link, will have to keep an eye on that one.

Cruisin,

Having tested one, I can confirm laptop power supplies are isolated. Although I absolutely agree that safety must not be compromised to save money, If properly fused on the input and output, I see little risk to either the pack or the car in trying this. As for what you are saying about items proved to work and sold by EV suppliers, this isn't always true. Take the vacuum pump kits as an example, they are very expensive and although they work, make an awful racket in use, The OEM VW/Audi/Landrover etc. vacuum pumps are far cheaper and run almost silently, I presume you prefer the former?!

This is a technical discussion forum to discuss matters such as this topic and help one another, I think your comment about getting attention is inappropriate. I come here for help, if you can give me a valid technical reason why this is a bad idea, I'll happily throw it out! Also, if this is "off the wall" may I remind you what we are all doing here!! Most of the population probably think everything we discuss here is "off the wall", by your logic, we may as well all give up and go and buy a Leaf!

DJBecker,

Thanks for the input, can you elaborate on the possible effect of paralleling these, regarding power output, would diodes on the +ve of all units help, although I suspect they already have one internally?

For what they cost, I might just buy four of them, parallel them up and apply 10A/20A/30A loads and soak test them at each for a few hours.

You may be interested in the customer of mine who came in 2 weeks ago with a dead $9k pack. He was using a home designed charger (to save a buck) and it failed taking out the entire pack. Yes, he could have installed some safety devices to prevent it, but that costs too. Cant tell if the charger caused the problem or the defective IOTA which isnt designed for the purpose of a converter like they are being used for. And I might add, the amount of failures as witnessed on this forum.
 
#44 ·
Tesseract, thankyou, that is very useful feedback. Allowing for this effect, what would you recommend as a most suitable DC/DC converter to use?

Cruisin, I completely agree that it is not worth risking the pack with a badly designed charger, either home brew or cheaper commercial type.

I still however can't see how a failing/failed DC/DC converter of any sort can harm the battery if it has a correct input fuse as any current it is taking that is inadequate to blow the fuse would be far too low to harm the battery. That is unless it was left connected all the time (very bad practice, surely?) and drained the battery dead on developing a fault.
 
#45 ·
...Allowing for this effect, what would you recommend as a most suitable DC/DC converter to use?
I don't recommend anything as none of the commonly used products are close to ideal for the job, including the Vicor models. That said, if you put at least 100uH of inductance (rated for the maximum dc current expected) in series with the input of the dc/dc converter, or the output of the charger, then that will give them a fighting chance. For example, this inductor from DigiKey would be a good choice for protecting all but the largest dc/dc converters, and smaller chargers and/or ones connected to higher voltage traction packs.

And, yeah, leaving the dc/dc converter connected all the time (you know, so that you can sacrifice your traction pack to protect your 12V battery from overdischarge :D ) is how a partially failed dc/dc converter can kill your pack. The device most likely to cause harm is the charger, however, as its output will likely be fused at a much higher current than the dc/dc converter's input.
 
#67 ·
No, indeed EvolveElectrics put their sticker on it so I don't know who actually made it. The price/performance did seem reasonable so I haven't spent any time figuring it out.

I am running it full-time with a backup aux 12V battery. It takes so little power to just be on, it's below the 100mV resolution of my voltmeter when I look at how the pack voltage changed over a couple days. And MTBF is rated to about 16 years always on. :)
 
#72 ·
Fits the specs at a good price point, and they warranty their product for 2 years. I'm not happy with how Evolve hides who made it, but it seems like all the EV suppliers are doing that. Other than that, what exactly is your problem with it?

I haven't been running it even a month yet, so I can't promise you it will make it to the 16 years it's spec'd at, but so far it is taking everything I throw at it.
 
#74 ·
Mean Well makes a decent, though not terribly robust, power supply. We use a bank of them to charge each AGM battery in our dyno. Every so often one blows up, and not because of abuse via the dyno, so we have learned to keep 4-5 on hand at all times. That said, at only $35 each for a 13.8V/11A power supply unit ("psu) *including* PFC, well... even my pride takes a back seat and just orders the damn things, rather than try to roll my own.

But let's be perfectly clear here: we installed the Mean Well psus inside a NEMA cabinet with filtration and fans to keep them clean and cool. The pc boards are not even remotely humidity resistant - CEM-4, which is literally a sandwich of paper and phenolic resin - and they run hot at rated load (not much safety factor in the design. They require a minimum load to prevent the output voltage from skyrocketing - and said load is provided by some power resistors on the pcb. These resistors continue to drain your battery (around 1-1.5A) whether the supply is on or not. A different topology choice and/or control scheme would have sidestepped this issue. There is also not a proper precharge circuit for handling DC supply voltage - just the usual NTC resistor. Cycling power to the unit before the NTC resistor has a chance to cool down again will blow something up sooner rather than later.

Etc... and so on. But hey, you can get 350W for $120 so who cares, right?
 
#75 ·
I'll start to care when this one dies, each one I have to replace means I'd be willing to spend that much more on a better one in the first place. Maybe after a few years you'll have a market for a $400 DC/DC for those of us who have to learn the hard way ;) It sounds as if the Mean Well design could be made better with some relatively simple fixes.

I've been trying to deal with some of the issues you brought up by disconnecting the input and output when the vehicle is off but I didn't consider the startup issue with the hot NTC resistor. Maybe I should add more cooling to help avoid that, what sort of times are we talking about for it to cool down, one minute, 10 minutes, more? Also is there a preferred timing of connection, output first, input first, or same time? I've been switching on the output first then the input, and disconnecting in reverse.
 
#76 ·
Hi just been reading with interest can I ask silly question.
Can you put 2 in parrellel to give you bigger output at same voltage . Plus if 1 fails you still have other to run emergency stuff to get you home. Am thinking of buying this unit so need clarification (the unit in question was the acme from evolve or evsource or whoever )

Cheers
 
#77 ·
Hi just been reading with interest can I ask silly question.
Can you put 2 in parrellel to give you bigger output at same voltage . Plus if 1 fails you still have other to run emergency stuff to get you home. Am thinking of buying this unit so need clarification (the unit in question was the acme from evolve or evsource or whoever )

Cheers
You may consider doing what I did for a customer who wanted the following and it has proven to work as intended. A Mean Well SD 100D-12 ($40) was installed to operate all the time keeping a 14.25v nominal 50ah Li-ion battery, made up from 18650 cells, charged. The Mean Well requires very little standby power, has no noisy fan, and if it fails, the battery is in parallel. The total 14.25v system is about 60ah which is more that what most people need. Less cells would lower the cost of the battery for a smaller system. The real advantage here is being able to use 14.25v for the system instead of 12v. I will let others add to this post to outline those advantages. It should be noted that the Mean Well has a voltage adjustment that could be used to increase the voltage to around 15v to keep the battery charged.
 
#78 ·
Cruisin,

A few posts ago on this thread you told of a customer of yours who had a $9000 battery ruined by a DC/DC converter. Therefore can I clarify your last post, you are now advocating leaving a small DC/DC converter permanently switched on?

I hope I've misunderstood, my advice to Spyder is to only have whatever DC/DC he chooses turned on when the ignition is on, sized appropriately to keep the auxillery battery topped up whilst using the expected loads as the car is in use.
 
#82 ·
Cruisin,

A few posts ago on this thread you told of a customer of yours who had a $9000 battery ruined by a DC/DC converter. Therefore can I clarify your last post, you are now advocating leaving a small DC/DC converter permanently switched on?

I hope I've misunderstood, my advice to Spyder is to only have whatever DC/DC he chooses turned on when the ignition is on, sized appropriately to keep the auxillery battery topped up whilst using the expected loads as the car is in use.
Because a car with a BMS and/or SOC draws 12v from the battery, some source needs to keep the battery charged while the car is not being used. My statement was that using the IOTA, which appears to have a high failure rate, one could have a problem if it fails. The samr problem exists with the Mean Well, but have not seen the failures as other DC/DC converters have experienced. Using the smallest available (100watt) will not put a big load on the pack while idle like the IOTA does. Using a seperate 12v charger to charge the 12v battery is useful but not part of a good design for many reasons.
 
#79 ·
What do you guys think of the Mean Well SD-1000H-12?

It is US$300 from JameCo 72-144vdc in, 11-15vdc, 60A out. It has a couple of interesting features; remote on/off, and 12V 0.25A auxiliary output. It has a 3 yr warranty, and lists a MTBF of 32k hours (3.65 yrs).

If the Aux 12v out is always on, it would be ideal for running the EV Works ZEVA Fuel Gauge Driver Plus, which requires an always on 12v, 35mA source.
 
#81 ·
What do you guys think of the Mean Well SD-1000H-12?
I like the 11-15 V output voltage range.

I don't like how they describe the overload protection: "Constant current limiting," (good) "unit will shut down output voltage after about 5 seconds." (not so good) "Re-power on to recover" (bad!)

I mean, if it's constant current limiting, it can withstand an "overload" indefinitely, right? It has separate overvoltage and overtemperature protections. Those I don't mind needing to restart to recover; in fact, I think it's a good idea for overvoltage.

I think it's just poor wording on the datasheet, but I don't like having to buy one to find out.
 
#83 ·
Hi cruisin,

OK, I understand your logic, but I think it's flawed. You are arguing that the 12v battery needs to be kept topped up whilst the car isn't in use or over time will need charging seperately via 12v charger. So in order avoid this you suggest leaving a parasitic load running off the main pack.

Lets take an example to show why this is such a bad idea. You can slate me for this later if I'm wrong! I'm always happy to have my thinking proven incorrect if it actually is. :)

Assuming a conversion has a typical 160Ah lifepo4 battery and you have your small DC/DC on it all the time, and lets assume you have a 2A fuse on the input to protect the main battery. Now lets say that the DC/DC develops a fault over a long weekend whilst the car is parked up and starts to drag a couple of amps constantly as a result from the main battery. We'll also take it the car has been charged and the charger subsequently unplugged for the safety of the pack.

Now 2A doesn't sound much does it?, but if left for only 3 1/2 days in that state the main battery that probably cost as much as the rest of the conversion put together is probably permanently dead! :eek:

Would it not be more correct to build using a correctly sized 12v battery that could last a couple of weeks minimum with the expected parasitic loads, as current ICE designs do and have a larger, suitably sized DC/DC converter capable of maintaining the necessary running loads and topping off the 12v battery whilst the car is in use only?

That way, there is no risk to the main pack, (at least from the DC/DC converter, as for a BMS left connected.... well that's an arguament for another time :p)
 
#84 ·
Hi cruisin,

OK, I understand your logic, but I think it's flawed. You are arguing that the 12v battery needs to be kept topped up whilst the car isn't in use or over time will need charging seperately via 12v charger. So in order avoid this you suggest leaving a parasitic load running off the main pack.

Lets take an example to show why this is such a bad idea. You can slate me for this later if I'm wrong! I'm always happy to have my thinking proven incorrect if it actually is. :)

Assuming a conversion has a typical 160Ah lifepo4 battery and you have your small DC/DC on it all the time, and lets assume you have a 2A fuse on the input to protect the main battery. Now lets say that the DC/DC develops a fault over a long weekend whilst the car is parked up and starts to drag a couple of amps constantly as a result from the main battery. We'll also take it the car has been charged and the charger subsequently unplugged for the safety of the pack.

Now 2A doesn't sound much does it?, but if left for only 3 1/2 days in that state the main battery that probably cost as much as the rest of the conversion put together is probably permanently dead! :eek:

Would it not be more correct to build using a correctly sized 12v battery that could last a couple of weeks minimum with the expected parasitic loads, as current ICE designs do and have a larger, suitably sized DC/DC converter capable of maintaining the necessary running loads and topping off the 12v battery whilst the car is in use only?

That way, there is no risk to the main pack, (at least from the DC/DC converter, as for a BMS left connected.... well that's an arguament for another time :p)
Your argument is valid, but I have not experienced a failure like you have used in your example. Of the 50 something conversions we have done, most use DC/DC 1/2 bricks as they are SS and the most dependable of all DC converters. Most people want to get away from heavy batteries and dont want to spend the money on Li-ion. We use 50ah Li-ion 14.25v nominal for for our cars. The charge does need to be mantained or a gradual discharge will take place when using a BMS and a SOC. If you have not used these in conversions, then you may not understand their use. By the way, hybrids use traction power to the inverter to provide static power for most uses other than lights, etc which is used by a small 12v battery. Also, using your example, you are stuck with 12v, which really doesnt do much for lights etc. How long do you think a properly sized battery will last with a 2amp draw at 12v? Figure it out.
 
#85 ·
I take your point that such failures as in my example may be rare, I just don't like the thought of any risk to such a large investment as a main battery.

I understand the use of BMS and SOC. Personally I don't like BMS and will not be using one, but I know that's another can of worms! (not BMS as in management anyway, monitoring, yes)

The 2A draw was an example of a fault condition, nothing like that will be taken from the 12V battery under normal parasitic loads, so a correctly sized battery should be fine for a while at least. If you meant a fault drawing power out from the 12v battery and killing it, well, better that than the main pack!! lol!!

As for the extra weight of a properly sized 12v battery, so another 10Kg to 15Kg tops, it's negligible in the scheme of things.

I also don't know why you keep referring to only having 12v, a properly sized DC/DC should be putting out 14.4v or so to maintain the running loads and charge the battery. (as an ICE alternator would).

At the end of the day, I suppose everything is subjective and there is more than one competent way of achieving the end result.
 
#86 · (Edited)
I take your point that such failures as in my example may be rare, I just don't like the thought of any risk to such a large investment as a main battery.

I understand the use of BMS and SOC. Personally I don't like BMS and will not be using one, but I know that's another can of worms! (not BMS as in management anyway, monitoring, yes)

The 2A draw was an example of a fault condition, nothing like that will be taken from the 12V battery under normal parasitic loads, so a correctly sized battery should be fine for a while at least. If you meant a fault drawing power out from the 12v battery and killing it, well, better that than the main pack!! lol!!

As for the extra weight of a properly sized 12v battery, so another 10Kg to 15Kg tops, it's negligible in the scheme of things.

I also don't know why you keep referring to only having 12v, a properly sized DC/DC should be putting out 14.4v or so to maintain the running loads and charge the battery. (as an ICE alternator would).

At the end of the day, I suppose everything is subjective and there is more than one competent way of achieving the end result.
I dont think there is a for sure solution to this problem as long as we have devices that have static drain on the 12v battery when unattended. I have tried various designs to lessen the possibility of a disaster, so far all my conversion customers have been happy. Does this mean it will never happen, NO. I did prevent all of them from ever having a DC/DC failure that could have stranded them or worse, damage the traction power pack. I keep refering to 12v because the standard car works on 12v, or about 13.5v with the altenator running. With a adjustable voltage output on the DC/DC converter, it is practical to boost it to about 14.25v. You cannot do that with a lead acid battery. I use Li-ion since 2p4s gives us 14.25v nominal with 5.4ah. Savings of a lot of weight as well. You are wrong if you think the weight savings is not important in a conversion. Its all about money and weight as to the outcome. Dont want to get into a debate on BMS here, but you are dead wrong on that issue. After building the first street legal Li-ion conversion in 2002, I have seen it all and would not drive a conversion with out a BMS or SOC, period.
 
#91 ·
How are you using the Mean Well's that blow up? Do you switch them off with the car? I've been trying to decide what the best way to switch them off is, input power off first, then output, reverse, doesn't matter? Tesseract says switching them off and then back on before the resistor cools down enough causes voltage spikes.
 
#92 ·
Killed them on the bench, not in the car. One I had running without enough cooling, the other was overloaded significantly. They work well but can be a bit fragile. I do have one that might make it in the car, and if I do it will stay on all the time. I have some bricks that might be used to power the stereo or other optional loads but they have a nice remote turn on that you can switch with a normal relay vs switching pack voltage.
 
#104 ·
My old Dodge had a mechanical regulator and I adjusted it so the peak voltage didn't exceed 15.0 volts.

I just checked 2 late model Toyotas in my driveway. One reads 14.29 volts at idle with everything off. The other reads 14.35 volts with everything off. These are factory stock, no charging system parts have been replaced. They even have the factory original 12 volt batteries.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top