Relay testing

[Bitsyncmaster]

FYI for you electrical guys.

I laid out a new fan relay circuit board (circuit change to prevent fault with open ground). So I decided to set up a high current test jig to decide if I should use 4 oz. copper instead of 2 oz. with the new circuit boards.

At 30 amps load current my fan relay rev B drops 0.064 volts across the power and load pins. That is 1.92 watts of heating.

Using the same test jig and installing a stock mechanical relay (rate marked as 30/40 amps) the drop is 0.120 volts. That is 3.6 watts of heating from the load. Add to that the heat produced by the coil current 2.45 watts and total heating is 6.05 watts. I wonder if all brands of relays dissipate that much heat at 30 amps. I know I've seen some marked at 50 amp ratings and think I've seen a 60 amp.

I think I may use 4 oz. copper when I place this next PCB order. It cost more since I have to order full production boards.

[Bitsyncmaster]

Some more testing.

Tyco 0-1432787-1
Voltage drop at 30 amps = 0.057 volts

Beuler BU-5083W-0000
Voltage drop at 30 amps = 0.056 volts

Tested that no name relay again
Voltage drop at 30 amps = 0.118 volts

So far it looks like you want a name brand relay for the high current circuits (fan relay or blower relays). Note that Beuler relay is marked 40/60 A. I think the Tyco is rated at 30 amp (not marked on case). These are all new relays so maybe with age the readings may get worse.

[Bitsyncmaster]

I tested the 70 amp rated relay I bought from Del City.

SONG CHUAN pn 897-1A-D1
Voltage drop at 30 amps is 0.043 volts.

I would expect about half of the voltage drop than the 30 amp relays. This one is less but not half. These do sound like minute voltage drops but when your current is very high they do produce quite a bit of heat.

[Exolis]

Neat. Have you considered using two 30 Amp Relays in Parallel? Typically vehicles nowadays with 2 Elec Radiator Fans uses 2 Relays in parallel, though I don't remember what the Relay ratings are, maybe 25-30A.

Otherwise for a 30A load, I would recommend 50A+ Continuous current, big relays like that don't have too much trouble withstanding a little bit of heat.

[Bitsyncmaster]

Well I placed an order for new circuit boards with 4 oz. copper. I had to buy 200 boards to get a decent price. Also had to use 370HR material which is higher heat and less thermal change then FR4. I also ordered better MOSFETs which just came out on the market. So these new relays should be rated to 70 amps. There are no super large terminals (used on the AUX) relays off the shelf. I've already dropped $1500 on this project so may not get those large terminals produced.

[Riley88]

This is awesome news as i am going to be buying some solid state relays from you very soon. Love what you do for this community!

[SS Spoiler]

Any luck on job prospects?

[Bitsyncmaster]

Any luck on job prospects?

Had one prospect but have not heard back for a few weeks. The company that bought us 18 months ago has let a few more people go. It is the age thing that makes getting a new job very hard to get. All of us over 60 still have no jobs. Suggest you all build your 401 K up as much as you can.

[tom kish vin. 4357]

Had one prospect but have not heard back for a few weeks. The company that bought us 18 months ago has let a few more people go. It is the age thing that makes getting a new job very hard to get. All of us over 60 still have no jobs. Suggest you all build your 401 K up as much as you can.

you got that right!!

[Bitsyncmaster]

Some more testing.

Tyco 0-1432787-1
Voltage drop at 30 amps = 0.057 volts

Beuler BU-5083W-0000
Voltage drop at 30 amps = 0.056 volts

Tested that no name relay again
Voltage drop at 30 amps = 0.118 volts

So far it looks like you want a name brand relay for the high current circuits (fan relay or blower relays). Note that Beuler relay is marked 40/60 A. I think the Tyco is rated at 30 amp (not marked on case). These are all new relays so maybe with age the readings may get worse.

I built one of the new relays that I ordered with 4 oz. copper. It also has the new MOSFETs which are 30% lower on resistance.

voltage drop at 30 amps = 0.032 volts.

[David T]

Had one prospect but have not heard back for a few weeks. The company that bought us 18 months ago has let a few more people go. It is the age thing that makes getting a new job very hard to get. All of us over 60 still have no jobs. Suggest you all build your 401 K up as much as you can.

The objective is to get the average age of the employees as low as you can. It can make a big difference with the health care costs. And when they hire new, younger employees to replace the older ones (if and as they may need to) they will tell them there is no pension. As for the 401K, get your money out of there as quick as you can and get it into your own IRA where you can control it. One thing your old company may do is to hire you back as they need you as a free-lancer. That way you are not on the payroll and there are no benefits for them to have to pay. No sick days, no vacation, etc but you do get to work and pay taxes! If you have skills that are still valuable the thing to do is to start your own company and market yourself as a consultant.

[Bitsyncmaster]

Here are some displays of a standard relay switching a resistive load (one headlight, 2.7 amps at 13.0 volts). Also attached is the same test with my new REV-C relay with the same test setup. I have purposely set my turn on and off times to slow the ramping of the load voltage to reduce or eliminate switching noise into an inductive load (Fans).

[Bitsyncmaster]

No luck finding the larger terminals used on the 70 amp AUX relays. I'm running the new relays with the standard size terminals. I guess you could make an adapter if you want to do the same. I just crimped new connectors on the OEM wires and used a relay socket. I don't expect much interest in solid state AUX relays since it only provides less heat and slower switching for less inductive noise.

I still have yet to test them above 30 amps. My bench power supply is limited to 30 amps. I could hook up a battery and charger for higher current. I also need more test load. I now use 6 headlamps wired to buss bars which gives me 30 amps at 13.7 volts.

[Chris 16409]

Dave, how is your new aux relay different from your originals that I'm using? I like mine just fine, just curious. Didn't you say it's based off of your fan relay board?

[Bitsyncmaster]

Dave, how is your new aux relay different from your originals that I'm using? I like mine just fine, just curious. Didn't you say it's based off of your fan relay board?

The original AUX relays you have were hand wired into the OEM cases. They had been working fine for the 6 or so years I ran them. What made me change them was the design of my battery cutoff relay. I needed to switch the AUX relay via the ground wire which those originals could not do. I designed a circuit change which lets the relays switch off when the ground is removed. Yes, all of my relays are based off the fan relay design.

[Chris 16409]

Oh I also installed that hot start relay for my friend. I told him to unplug his thermotime time switch as well. Works great. He likes the piece of mind that his cold start valve will fire if there is a hot start issue.

[Bitsyncmaster]

Oh I also installed that hot start relay for my friend. I told him to unplug his thermotime time switch as well. Works great. He likes the piece of mind that his cold start valve will fire if there is a hot start issue.

Always glad to hear of another happy customer. Saves the embarrassment of doing the plug swap.

[Bitsyncmaster]

It's been bugging me that I have not been able to test relays above a 30 amp load. 30 amps is the limit of my bench power supply and my load consisted of 6 headlamps.

I'm building a new resistive load with 32 six ohm 50 watt resistors. Each resistor will provide a 2 amp load at 12.0 volts so all 32 with load 64 amps. I'm setting up 7 switches so I can select loads from 0 to 64 amps in 2 amp steps.

Now to power my test circuit I have a car battery that I plan to attach high power battery charger onto during testing.

For long high current testing I may have to have water running over my load plate. I'm also installing a 0.01 ohm,100 watt current shunt that will provide a current reading of 1% accuracy. I may be able to test relays to failure. I have a few standard relays that claim 60 amps and I know measuring the voltage drop on those that they will never work at 60 amps continuous load.

[Bitsyncmaster]

Not that anyone else will ever need to build one of these.

This load tester is a lot of work. Over a 100 holes drilled. Now all I need to do is mount the "switch plate" and run 8 wires from the resistor plate to the switch plate. I wired it all with 16 AWG buss wire and used Teflon tubing for insulation. I connected those wires so each should not have to conduct more than 10 amps. I need this load tester for testing my products but it could be used to capacity check batteries also.

I've been reading up on lead acid batteries. You load test so a fully charged battery drops to 10.5 volts in 20 hours. That is how the amp. hour rating is specified. Also discovered that it's not bad for a lead acid battery to be fully discharged if you immediately charge it back up. It's lead sulfation that forms over time with a discharged battery.

[Bitsyncmaster]

The test set is done but now I'm making up cables to hook it up to a car battery. My bench supply will only provide 30 amps and I want to do testing with higher currents.

The test set is 24" x 12" made with 0.160" 6061 aluminum. Even with that large heat sink, I will need to run a fan blowing over it for testing longer than 60 seconds.

I did temporarily hook it up to my bench supply. I was impressed my bench supply current meter reads exatly 30.0 amps when the test set current monitor (a 0.01 ohm 100 watt 1% accuracy) also read 30.0 amps.

My cabling will consist of six 16 AWG wires for each power, ground and load connections. That group of six wires results in almost an 8 AWG wire.

I set up the loads as six separate loads so I can use this also for testing my Fan Fail units (those need two loads). For high power relay testing, I just connect all 6 loads together.

[Bitsyncmaster]

Ran a quick test with my new test set and a rev C relay. I switched to max current and was getting 58 amps. The Rev C relay stayed cool but the connector pins started to get hot. I let it run for a few minuets and then shut it down. The resistor plate on my test set was hot. I did not have a fan running over that plate yet.

So it looks like my weak point is the connector pins (think it's the female "wire" side that producing heat).

The pins I'm using are made with 0.013" metal stock. I looked for some thicker stock pins and found some made with 0.018" stock. I placed an order for 100 each of those pins. One is rated for 8 to 10 AWG wire and another is rated for 10 to 12 AWG wire. The wire size is just how long they make crimp tabs. Wire size has nothing to do with power rating although thicker wire will carry away more heat.

[Bitsyncmaster]

Well I goth those heavy duty terminals for the relay pins. Those things really take to much force to remove from the relay pin to use in relay sockets. I guess you could use them if you plug and unplug one pin at a time (no socket). But they worked out fine for my high current testing. The pins get a little warm at 56 amps but that may be heat generated by the relay internal traces.

A blower aimed at my resistor heat-sink plate keeps it only running warm to the touch.

Anyway, I can see how we get some relay sockets and pins melting. Any corrosion or weakness in the pin and crimp will get hot running high currents (30 amps).

There also are better grade socket pins. The FASTON brands can be bought in a premium grade. That premium puts the pins through a stress relief process to control the pressure the contacts supply.

[Bitsyncmaster]

I ran across this application note paper for car relays. It is interesting the use of using PWM to power the relay coil to reduce heating produced by the relays. I thought I was the only one concerned with relays producing heat.