A review – The Five Fish EL-2 Electronic Load
At times a constant current load is useful on the test bench. It can be used to test a power supply, to characterize battery performance and for a host of other things. I’d been looking to design and build my own constant current load and had started collecting parts for one. I came across the Five Fish EL-2 Electronic Load during an eBay search and decided to order the blank PCB.
Note that I’ve included some assembly notes towards the end of this blog.
A few days after ordering the PCB it arrived. I went through my parts stock to see what I already had on hand then hit Digikey to order what I didn’t have on hand. Five Fish provides a parts list on Digikey. While I waited for the parts to arrive I went by my local computer recycle center. They tend to have a large selection of old computer and electronics items included user computer heat sinks with fans. I picked up three identical heat sinks with fans so I’d have a spare should one have a failed fan. I paid about $2.00 for each one so having a spare was cheap insurance.
The build was very straight forward requiring typical electronics assembly tools. I should have documented the assembly process but didn’t. I dug through my hardware box for spaces for the fans and feet on the PCB.
I used a power switch from my junk box along with an old BNC connector for the oscilloscope monitor output. They may not look as nice as the suggested parts but certainly get the job done.
With the unit assembled I went on to testing it. It requires an external 12VDC power supply, a 12VDC wall wart or other suitable adapter will work. My assembled unit draws about 150ma so pretty much any 12VDC wall wart would work.
I went on to load testing the unit.
My test setup:
- Rigol DP811 Programmable Power Supply – capable of 40 volts at 5 Amps or 20 volts at 10 Amps. The DP811 can deliver 200 watts making it ideal to test the EL-2.
- Rogol DM3058E bench DMM.
The EL-2 was connected to the DP811 power supply. The EL-2 provides an output that can be used to monitor the current being sinked, it outputs one millivolt per milliamp being synced.
I enabled the output on the DP811, set it to 40 volts with a current limit of 5 amps. I adjusted the current draw using the 10 turn potentiometer to obtain 4.51 amps current draw (180 watts). The DP811 has a digital current meter with 10 milliamps resolution. The current meter on the DP811 tracked very well with the ‘current’ reading my DM3058E was displaying from the current sense output on the EL-2; this output works very well and tracked very well with the DP811.
Rigol DP811 showing 4.51 Amps current draw / 180.40 watts power dissipation.
Rogol DM3058E measuring the ‘current’ output of the EL-2 showing 4.5090 amps current draw – the current sense output on the EL-2 is spot on.
I left the EL-2 sinking the full 180 watts for several minutes. The FETs rose to about 95°C and stabilized there. The current draw stayed right at the 4.51 amp programmed value as the temperature rose. The feedback design on the EL-2 looks very solid.
The only thing that concerned me a bit was the row of 2 ohm resistors used to create a 0.2 ohm current shunt. These resistors rose to about 120°C. If you decide to build you own EL-2 go ahead and buy 1/2 WATT resistors then install them raised off the PCB, being raised off the PCB will allow better air flow around them. When I make my next Digikey order I’ll be replacing the 1/4 WATT resistors I bought with 1/2 WATT resistors and will mount them with an air gap.
Assembly was straight forward, the assembly manual is laid out well and was easy to follow. Of course being a bare PCB you’ll need to order your own parts and heatsinks. You’ll also need some hardware to mount the heatsinks.
Two of the challenges in assembly of the EL-2 are the heatsinks and mounting the MOSFETs.
I was surprised how difficult it was to drill the screw holes in the heatsinks to mount the MOSFETs. I broke off a drill bit in the first heatsink I drilled into. Although the material looks like copper it must be an alloy based on the effort it took to drill the holes. Take your time and let the drill bit do the work. It took quite a bit of time to finish the holes. As I don’t have a tap and die set I approached this by using self taping screws. Given the hardness of the material using self taping screws was also a bit of a battle. After making sure the screws I had fit through the mounting holes in the MOSFETs I drilled a few holes in a piece of scrap wood to find a bit that seemed well sized for the screws; not so big the screws were loose in the holes and not so small the screws wouldn’t be able to bite in and self tap the hole. After drilling the holes I ended up enlarging them a bit with a bigger drill bit so the screws could get a bite. I used a larger drill bit to countersink the holes to remove the burr left over from drilling the holes.
One of the considerations in mounting the MOSFETs is providing strain release for the them. I used wire from an old floppy drive power cable which I drilled to the MOSFET’s pins. I slid each pin partway under the wire’s insulation then soldered the wires to the pins. Heat-shrink tubing was then used to cover the solder joints. You may notice in the photo below that the wires are solder to the EL-2 PCB at an angle. By having the wire some out of the board at an angle it will help keep the diameter of the bend in the wire as large as possible. This will help reduce stress on the MOSFET pins.
Be sure to read the section in the assembly manual about insulating the MOSFETs from the heat-sinks.
With the heatsink mounted to the PCB the wires bend fairly smoothly to the MOSFET. Being able to hinge the heatsink on the wire as shown above will help should they even need to be replaced.