Solid state relay problem.

[bigrockcandymountain]

Today i tried hooking up my diversion circuit using waste not high pwm on the classic controller.  That part seemed ok but the settings will need to be played with. 

So i hooked up a 1500w 48v water heater element and a 60a dc ssr from china on heat sink (too small a heat sink). 

So while it was working voltage drop across the ssr was 2.5volts.  The element has slightly lower resistance than it should like 1.4 ohms and my voltage was 60v which is normal absorb.  So questions...

Who has a good brand dc- dc ssr and what is the voltage drop across it?
Will it hurt a 1500w element to run at 2500w? (60v @1.4 ohms )
Can i run a small brushless dc fan off the output of the ssr for heat sink cooling?

Any other ideas what i should do would be great too. 

Thanks
Derek

[OperaHouse]

PWMing many DC SSR bricks can be a problem. Because of the slow response there can be a lot of transition heating. Some have a overhead voltage that it can never drop below. No real experience because I build rather than buy. Small fans can even be run with PWM if isolated with diode and filtering cap.  Many ELECTROBIC wall warts operate on 60V and they could be used to power the fan.  A 1500W is already too high a heat density, it can be run at max power if pulsed to that power level.

[bigrockcandymountain]

Thanks for your reply operahouse.  I am a little confused about pulsing to max power the element. Could you explain that a bit more? This element is actually two in parallel mounted on the same base. 

I will get a better relay.  The specs say 0.4v max voltage drop for the mosfet output crydom brand units. 

Sizing the heat sink i could use help with.  Say 30a at .4 volts so 12w. Ambient is under 30 degrees celsius.  I would rather have convection than a fan. 

[kitestrings]

We're pre-heating water too.  We've been using SSR  from VB Controls, Virginia:
http://www.vb-controls.com/index.html

and the guy that I've corresponded with is Steve.  Very helpful.

We had used some inexpensive ones that over time just didn't hold up.  We have large heat sinks, with grease.  Ours are about 4" x 6" with 1" fins and they are mounted back to back for two relays.  We also added an MOV or flyback diode as per his recommendation.  No fan; but it is in a pretty cool space (~55 degF).

~ks

[kitestrings]

Sorry I ran out of time there...here's an early picture of ours.  I since have remounted it so that the fins are vertical which aids in cooling quite a bit.


You asked:

Will it hurt a 1500w element to run at 2500w? (60v @1.4 ohms )

I'm not sure I understand what you have.  Certainly if you run something rated for 1,500w at 2,500w it will burn out, eventually if not immediately.  Most of the DC elements that I've seen are actually dual elements with a little copper shunt bar to configure them as you wish.  One I'm familiar with is a 24/48V nominal element; intended to be driven/dumping at about 29V & 58V respectively.  It can be wired for 22 or 48A at 24V, or 22A at 48V.  So if you do the math here, there are two 1.24 ohm elements.  They can be wired in series or parallel.  At 48V then, you would get (58/2.48ohms)= 23A, or 1,356w, but you can't double this without a different resistance element.

Hope this helps.  ~ks

[bigrockcandymountain]

Here is a picture of the element.  It has two separate loops with the removable bars.  When the two are hooked parallel it tests 1.5 ohms.  That is 1536w at 48v and i think what it is rated for. That is how i had it hooked up (2400w at 60 volts or something like that). The loops are 2.8 ohms each so i could hook only one and have 1285w at 60v i think it would still burn out fast. 

Series is 5.5 ohms so very small power.  It would be very safe though. 

I will look in to those ssr that you suggested.  I read a quick bit on them. They are faster switching i think than the crydom i was looking at. 

A larger heat sink and I'll be on my way. 

[DanG]

Watts (power, work) remains the same while I & E vary; the OEM stamping on that heater is an absolute rating, yes sure is can be pushed harder example heating running water but runs risk of damage to the seals or wiring terminals blah blah if it remains energized in still water, or go full china syndrome if it goes dry instead of merely opening like a fuse...

These are nominal numbers...
1500w @ 48V = 31.25A
1500w @ 60V = 25A
1500w @ 24V = 62.5A
1500w @ 12V = 125A

If we are pushing charge voltages (58.8V for 48V bank) then the amperes need to be scaled back from 48V & 31.25A to 58.8V & 25.5A...

Anyhow - yes anything can be 'overclocked' but Murphy's Law says we'd lose that heater the moment it was most needed.

[kitestrings]

I think you will find it is intended to be +/- 1,500w, and the math is like this, with nominal voltages (48V, 24V respectively):
   R   R   R_Tot   I   P
series   0.77   0.77   1.54   31.25    1,500
parallel   1.30   1.30   0.38   62.50    1,500

It can be hard to get an accurate resistance reading with an ohm meter, and the resistance can change after it heats up.  You're better to measure the current, measure the voltage and do the math if possible.  I'm doubtful that you have 5.5 ohms.

If you power only one (half) element, it will likely burn out.

[joestue]

how long it will last at 2400 watts rather than the 1500 watts it was rated depends on how close to the limit they already designed it. i suspect most stove heating elements which run at 2600 watts per 5 feet of 1/4th inch diameter rod (glows at about 800C at this wattage) will last a long time at 5000 watts if the outer jacket was cooled with hot water.

the power density for hot water heating elements may be limited by practical limits for noise, calcium buildup, and other concerns. depending on how cheap they are i'd try it but make sure you have a spare.

[bigrockcandymountain]

Thanks guys.  When i get back at this i will try it hooked series.  If that is way low i will just run it like i had it.  I don't have a great ammeter but when the batteries are floating you can pretty much tell by the classic display how many watts are going out.  It was fairly cheap so burning it out isn't too sad. I am not completely opposed to 3 or 4 2000w @ 120v elements that would be 500w @ 60v.  Im trying an unpressurized tote for hot water storage with an immersion coil heat exchanger.  Im already wishing i had gone the normal water heater route.  Our hard water eats them out pretty fast.  That is why I didn't. I like the marathon water heater kitestrings uses.  That would be a great option here. 

[OperaHouse]

I use those 2,000W elements on my 60V system. My Best Wind Dump Load would work well with those.  Too bad no one can work with electronics as they are only $5 each.  All that heat loss in a SSR drives me crazy when $5 for a half dozen FET would run cool.

[bigrockcandymountain]

Operahouse ya the ssr thing drives me a little crazy too.  They don't seem ideal but i think i will have to go that way for now. 

As a side question...where do I start learning about electronics?  I am amazed at the potential but don't know where to start.

I tested the element i have and it draws about 500w hooked in series which agrees with my 5.5 ohms and draws more than 1900w hooked parallel.  ( panels were only putting out 2000w. )  This was done @ 54v which is float. 

I will just run it at 2400w until it fails and then rethink from there. 

I contacted VB controls about their ssr and what they recommend.  Tomorrow is monday so hopefully hear back soon. 

[JW]

Here is a link to one for my posts, I bought a book on this topic, but cant find it at the moment.

https://www.fieldlines.com/index.php?topic=146533.0

Ya defiantly this link, must login to see it.

[JW]

Here it is

Some digging in that post should help

https://www.ebay.com/itm/GE-SCR-Manual-2nd-Edition-1961/142769228467?_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%26asc%3D50998%26meid%3Da31d66ffa1aa494cba4b71ee7cfd6807%26pid%3D100011%26rk%3D3%26rkt%3D5%26sd%3D202270589577%26itm%3D142769228467&_trksid=p2047675.c100011.m1850

[OperaHouse]

FET are easy as long as you switch below 10khz.  Put five FET in parallel and put 10V on the gate with a 2K resistor tied to the source. Each FET peak pulse current should be able to handle the rated amperage.

YOU learn by doing and burning stuff up.  Education always costs and you will never believe something someone just tells you.  I'v never posted anything a 15 year old couldn't do. They are adventurous and don't understand NO. You are the only one saying you csn't do something.

I sent off a working module to someone that seemed capable.  Only six wires to connect. It still isn't working. I figure anyone can connect one wire a week with under 30 emails for each one. Frankly, I give up.

[bigrockcandymountain]

Hey thank you all.  I had a laugh that the manual was written in the sixties. That is exactly what i want though. 

[bigrockcandymountain]

Sorry to be full of questions.  What model FET would you use?  Where do you all buy components? 

[OperaHouse]

There are two cheap FET I buy a lot of.  Search online for data sheet and ebay for 75N75 and 110N15.  They may have prefixes to designate manufacturer.  Just because something says 75A does not mean it can be used at 75A.  The leads are smaller than most fuses.  On resistance is key and FET in parallel lower that. They will share current like your brother, not evenly.  Adding a little wire length to each FET encourages them to share a little more.