Exceeding maximum voltage input on a DC DC converter.

[Circuit Learner]

I found a perfect DC-DC converter for my application, but it is rated for 36-75 volts input voltage.

I will use a lipo battery pack rated at max charge of 82 volts. The voltage will range from 66 volts to 82 volts as I use the battery.

Do you think I can get away with exceeding the maximum continuous voltage rating with this Dc-DC converter?

It won't be a big deal if it blows or burns up, so I am willing to take the risk if you think it will probably work.

Here is link to item.

https://www.mouser.com/ProductDetai...hrYxhSudp5mjKxZZHbIPAXYe8qzi31tVAS/70fEn6bw==

Data sheet link.

**broken link removed**

- - - Updated - - -

The DC-DC converter above is rated at 25 amps. I will only use a maximum of 10 amps and probably lower if that has any significance.

 

[BradtheRad]

No one can say for sure "Yes it's okay to use". (Nor does anyone want to be held responsible).

On the plus side, Mouser is a reputable mail order house. It's in their interest to stock only merchandise which has not been much problem to their customers.

Question for you to consider: What factors could there be to permit, or not permit, greater than 75V? Is it design sense, or does it really have a component that will blow up at 76V?

Speaking for myself, if I were constructing such a power supply, it would be silly for me to make it so it blows up at 76, or 77, or 78V. I would give it a large headroom, say 20 or 30V, to allow for component variations, mistakes in usage, spikes from unknown causes, etc. (Notice the datasheet specifies a 100V transient spec for a brief time.)

This doesn't mean the manufacturer gave it sufficient headroom to use at 82V. It's up to you whether to gamble (51 dollars) on how much much manufacturing sense the unit has built into it.

In case it were to fail and draw heavy current, you need to put a fuse inline, to save your Lipo battery.

 

[treez]

You should find a PSU that meets your spec...there are loads of them, literally loads for you...
try
digikey, Newark, RS, farnell as well as mouser.
Also, meanwell, Recom etc etc

try "open frame power supply"
"open frame switch mode power supply"
"switch mode power supply"
even just "power supply"
etc etc

 

[Circuit Learner]

You should find a PSU that meets your spec...there are loads of them, literally loads for you...
try
digikey, Newark, RS, farnell as well as mouser.
Also, meanwell, Recom etc etc

try "open frame power supply"
"open frame switch mode power supply"
"switch mode power supply"
even just "power supply"
etc etc

thanks for the recommendation.

For my requirements this will go on an electric bicycle and weight and space is very important.

You can not beat these tiny telecom power supplies for price/performance/weight/size.

It seems there are very few power supplies that supply higher DC volts and are small, and the ones that do are very expensive.

These telecom power supplies are so small that i can actually install three of them into my motor controller case, to give me 5 volts, 8 volts, and 12 volts. That is just amazing. Weight will be around 150 grams for all three. And the performance is amazing at around 20 amps each.

The only drawback is that I will exceed the voltage limits. Normally these are rated up to 80 volts absolute maximum, with a range of 36-75. I think delta just never updated the specification for the new model to include the 80 volts maximum. I would only exceed this by 2 volts. Around 95% of my use will be under 80 volts.

 

[schmitt trigger]

You have found the perfect DC-DC converter for your application, with the exception that no, you really have not.

On an electrical vehicle, whether an auto, a forklift or a bicycle, if an intermittent connection appears during regenerative mode, what is called a "load dump" situation occurs.
The voltage will skyrocket well above even the maximum normal. 2 to 3 times above nominal maximum is the industry standard.

The converter you choose should be derated for at least 20% above your continuous maximum, with transient capabilities at least twice that.

You may counter with: "hey I've tested it and it works fine" response. And my question will be, yes but for how long? Will the converter protect itself and shutdown, or will the magic smoke come out?

 

[Circuit Learner]

You have found the perfect DC-DC converter for your application, with the exception that no, you really have not.

On an electrical vehicle, whether an auto, a forklift or a bicycle, if an intermittent connection appears during regenerative mode, what is called a "load dump" situation occurs.
The voltage will skyrocket well above even the maximum normal. 2 to 3 times above nominal maximum is the industry standard.

The converter you choose should be derated for at least 20% above your continuous maximum, with transient capabilities at least twice that.

You may counter with: "hey I've tested it and it works fine" response. And my question will be, yes but for how long? Will the converter protect itself and shutdown, or will the magic smoke come out?

I'm not happy about having to exceed the specifications. However, there really doesn't exist anything out there as small and as powerful that will meet my voltage requirements. These seem to be cost effective because they are made in large quantities as a standard size and power.

Take a look at the mess I want to avoid on my new bike build. Below you can see the power supplies for +12 volt, + 8 volt, and +5 volt. I don't want all that mess anymore. The power supplies I want are so small they can be placed inside my motor controller and use the case as the heatsink.

 

[treez]

you could always have a look inside and see if stuff in there will be overvoltaged by the 82v.
Plug it in to 82v, put goggles on, put it behind a safety screen and see if it blows.
See if anything gets wildly hot.
Maybe there is a startup regulator that gets overheating with constant 82vin.
Maybe there is an overvoltage protector with eg a TVS that will get pranged by the 82v.
Maybe the FET isnt rated for the 82v.
What topology is it, if its a bridge then its fets would need rating to at least the vin.

- - - Updated - - -

You could always put in some kind of ovrvoltage switch which just literally switches the power supply out if the vin goes above 75V.
-Switch activated by a comparator and rseistive divider, and with a tvs the other side of the switch (downstream)

- - - Updated - - -

These guys may have something for you...they do those "just a bit higher voltage" power supplies...

http://www.amtex.com.au/

 

[schmitt trigger]

I hear you loud and clear....

If you are willing to add a small module ahead of the unit, you can always use a linear pre-regulator.

You can build a "power zener" with a properly heatsunk NPN transistor, and a zener. To drop a constant 7 volts at the input.

 

[Circuit Learner]

I have major problems. However, I don't think anything is wrong after looking at the specs.

When I connected 73 volts to my DC DC isolated power supply, I am able to get 12 volts on the output.

However, the power supply gets very warm and it has the heat spreader. It keeps climbing and I noticed temps were at 130F and seemed to continue climbing. The unit was getting very hot.

I used my fluke to check the amps on the unit.

WIth no load turned on it read 90 mA. With unit off it read 15 mA. Input voltage 73 volts.

Looking at the specifications they give the no load input current between 80 and 140 mA. SO it seems to be in spec with m y 90 mA reading.

Off input current they say range from 10 to 26, mine is 15 mA, so this is also in spec.

Is this normal to use that much power when there is no load? How do I deal with all the heat?

DId I buy the wrong power supply? WHy does it draw so much current and get so hot when in idle?

Thanks

Here is the datasheet and you can look at the specs.
https://www.mouser.com/ds/2/632/DS_Q48SC12025-372467.pdf

 

[treez]

yes its normal for a PSU to draw about a watt in idle, its because it maybe has a linear regulator to supply bias current to the control circuitry.
Expensive psu's use more speical circuitry to cut down no load power.

- - - Updated - - -

it says vin=75vdc Absolute maximum, so i wouldnt go above that. (ok it says you can go up to 100vdc but only for 0.1 second)

 

[Circuit Learner]

yes its normal for a PSU to draw about a watt in idle, its because it maybe has a linear regulator to supply bias current to the control circuitry.
Expensive psu's use more speical circuitry to cut down no load power.

- - - Updated - - -

it says vin=75vdc Absolute maximum, so i wouldnt go above that. (ok it says you can go up to 100vdc but only for 0.1 second)

Thanks

This was an expensive PSU. $50 dollars.

I would think a cheap one are the china ones.

The unit is tiny at 60 grams.

 

[Circuit Learner]

Very bad news, I decided to test it on 82 volts and guess what. It never turned on and won't turn on even at 50 volts.

I don't think it is a fault that needs to be reset because I assume disconnecting power and reconnecting power again would restart it.

Do you think I burnt it out immediately connecting it to 82 volts?

Well, lesson learned. I'll try and justify the loss to learning a lot more about power supplies with the hours of research I did for this.

Going back to my cheap china heavier power supplies. They just seem to work and work even with over volting.

 

[BradtheRad]

Did you have a load attached when you hooked up 82V?

Let it sit for a while, grounded to earth in case it helps to let all circuits fall back to 0V. Also in case ESD (or internal protection from ESD) is a cause in any manner.

Later try hooking up 50V again, with various loads attached.

 

[treez]

we'd need the schem of the psu to see if it could manage 82v, or be able to be modified simply to manage that. The chance of us getting the schem, as you know, is nearly zero.

 

[Circuit Learner]

I'll try to see it if will just work again, was trying that yesterday.

I'm hoping I can return it for my money back.

What surprises me is that the default over voltage value is 110 volts. I really thought it would handle the 82 volts.

The only thing I can think of is I didn't precharge it properly with a resistor before connecting power and had a large spark the 2nd time I connected it. I had a smaller spark the first time and it still worked. I doubt not precharging it would have blown it like that.

 

[treez]

it might have latching overvoltage shut down if it sees >75v for more than a second or two.

- - - Updated - - -

i fi was there i would look for component markings so i could check voltage ratings etc...capacitor voltages etc etc

- - - Updated - - -

i wonder if there is a startup regulator and that is under threat from the >75v.
Of course, as you know, overvoltage could mean the transformer going more into saturation....though as you say, there usually some margin designed in

 

[FvM]

What surprises me is that the default over voltage value is 110 volts. I really thought it would handle the 82 volts.

You are referring to the SM-Bus parameter? The datasheet (DS_Q48SC12025_06022015) however says absolute maximum rating 100V for 100 ms. Did the module ever work with rated voltage? Presume you wired the enable input.

 

[Circuit Learner]

Yes it did work at 73 volts. Yes I wired enabled, but even without enabled on it still draws a small amount of current.

It has hard to imagine ESD damage unless it can happen on the big voltage prongs as that is all I have touched.

It seems the 82 volts broke it. Never even turned on for a second. I had my fluke connected to it to read the amps and the amps never registered anything I saw.

I think when you have these tiny power supplies with all those smd chips on them they are very easily broken. Those cheap china power supplies seems to fare better with reliability.