This is the first time I've mounted the power supply up in the 19-inch rack that will house both the power supply and the four-pallet BLF188xr *water-cooled* amp. That's right, water cooled...oh, I didn't mention that to you before? :-) Yeah...I bought some very nice water-cooling plates for the next phase in this project. Anyway...I digress. So here's a few pictures of the control panel with the power supply mounted just below it in it's permanent 19" rack home.
(yeah, excuse the mess on the floor :-)
When you do this much work
in a short period of time things get a little messy :-)
The next thing I need to do is clean up this wiring and get the Ammeter wires re-connected also.
I really wanted to get all of this mounted in a box where it wouldn't extend out the back like that. But this was the least expensive option I had at this point. Most likely I'll build a box in the next revision of the power supply. Because I really would like to have it all fit in the rack front to back.
But for now this is all working great. And I can live with the way it looks.
The next phase is to order the RF amp parts, and start building that. I don't think it take nearly as long as this power supply did. I'm glad I started this 'winter project' when I did.
A few weeks back (maybe a month ago - not sure) a friend of mine asked me how the RFI/EMI was with these Switch Mode power supplies I'm using. At the time based on my tests at that time I hadn't noticed any RFI on my EME radio. About a week or so ago I was finishing up after a load test I was doing and I happened to notice when I turned the power supply on, that my EME radio had been on, with the volume turned up and when I turned the sspa power supply off I suddenly heard the noise floor drop quite a bit. After a LOT of research and poking around (literally) inside and outside the SSPA Power supply I've been building I finally decide to test for RFI using one of my spare power supplies that go inside the SSPA supply. I had already prepped it to go in the SSPA supply in case I ever needed to replace a failed one, so it was just like the others in the project. I decided to do some tests (this is pretty much all I could do to try to track this down at this point without an oscilloscope handy).
use my ears with the EME radio set to 6khz wide USB with AGC OFF and noise reduction and DSP turned OFF.
visually watch the S-meter (when I turned the SSPA supply on/off.
use WSJT 10's "measure" function to measure any changes to the noise floor.
use my HackRF One and SDR# (software) with an RF Probe I made from a piece of coax.
This supply I would test had one of the protection diodes across the positive and negative terminals that I've previous written about. One of the first things I did was test this spare supply with and without that diode attached. It turned out that the major source of RFI I was noticing was directly related to this diode. With it on, the band was full of noise, and off it would almost disappear.
So my next step was to remove all of those diodes. Once that was done I went back to tracking down the RFI that was remaining. At this point I set the spare supply aside and went back to working on the completed SSPA Supply with 8 of them inside it.
I installed 21 material 43 Fair Rite snap on ferrite beads on all the DC wires, as well as the AC wires inside the case. This made a significant difference. It got my noticable noise on the EME radio down to being only about +1dB when I turned the power supply on.
(WSJT - measure function shows RFI at this point)
(note also the copper wire I'm using to ground the cases to Earth Ground on the large case)
The 12vdc source I use for control of various circuits inside the SSPA supply was running on 120vac on one side of the 240vac AFTER the EMI inlet filter. I changed this over to run on 240v so that it would draw evenly on both sides of the 240vac. I read that you shouldn't run something on one side of the 240vac if you are using an EMI filter because it sets it off-balance for lack of a seriously in-depth description. I also replaced a 12 inch long Earth ground strap to this supply with about a 2 inch strap that runs to the SSPA supply case (which is Earth Grounded). I added some capacitors across the terminals of the 200 amp 60vdc Final output 'contactor', because I was seeing a spike in noise when this was energized.
(also note the ferrite [left] added on the on/off control line for the 12vdc supply [right])
These changes got my RFI down to <= 0.5dB (measurable, and repeatable) on the EME radio. The final step was to add capacitors on the 13.5 vdc supplies inside the SSPA supply. On all 8 of the supplies. I used 3.3uF and 4.7uF caps from the floating supplies Negative terminals to Earth Ground. And I used 2200uF Low ESR electrolytic caps across the Positive and Negative terminals of each of the supplies.
After this I no longer can detect any RFI on the EME radio when the SSPA Supply is turned ON.
There IS however, RFI *INSIDE* the SSPA Supply case.
(this was using my HackRF One and an RF Probe I built with the Probe INSIDE the SSPA case)
But it's no longer radiating out of the case. So there may be more work to do, but at this point I'm calling it resolved.
(OUTSIDE the SSPA case - no sign of the RFI just some other local/internal birdies)
So I *WAS* seeing S5-6 noise when this all started. Now there's NOTHING on the S-meter when I turn on the SSPA Supply. And WSJT 10 'measure' function can't detect any changes in the noise floor when I power it on/off. The HackRF can still spot RFI inside the case, but just outside the case it no longer does. And my ears don't detect it at all.
PS I would like to re-install the protection diodes again at some point and I may attempt that. I've since discovered that these diodes are probably ringing and I probably require some Fast "SOFT" recovery diodes instead. I've also discovered that some folks add caps around the diodes on the same terminals. It might be that the caps I have on mine noise could handle the noise from the diodes. But for now I'm gonna leave it all alone until I can do another big load test. (which may happen this weekend).
Success! Load test #6 is probably the final load test that I'll need to do with this power supply. I had no issues with the power supply. The final protection diodes didn't get much hotter than 95F during this test of 62 amps @ 53 vdc. In the last test I measured peaks around 180F. So my junk-pile heat sinks and muffin fans cut their temps about in half.
This test proved that my fixes for:
The high temps on the final diodes reduced to 95F which is 50% lower than what they were at, and only 33% of their max temp spec. Plenty of headroom there now.
The new contactor and time-delay resolved sporadic initial startup and immediate power down of random supplies (safe mode due to surge).
Replacement of the 400A shunt with a 150A shunt resolved the low amp reading on the 150A meter (doh!). I'd read that I should overrate the shunt for safety, that proved to be incorrect. The ammeter now reads accurately.
The voltage monitors work in-circuit now. If voltage is below or above range, the final contactor is forced open.
The only issues I had with this test was that the wires (#14 ga) on the Load Tester elements (and not the power supply being tested) started to get too hot after about 6 minutes of testing. THAT is acceptable for now :-) But I'll get some #10 for the future I think. Maybe even #8 for safety. If those wires had melted through I'd have had a fire hazard.
Fairly LONG demo video (not real stable if you get sea sick take your meds now LOL).
Here are the 'fixes' (or changes) post Load Test #5. I've added a 200A final contactor so that I can use the voltage monitors to control if we allow power output from the power supply based on voltage being in the proper range. I can also now delay allowing voltage output until all the supplies have stabilized on initial startup.
I've also added fans and heat sinks for the final dual diodes.
Here's a short video demo of the smoke-test after these changes were made.
As mentioned previously on this blog one of the issues noticed during the most recent load testing of this power supply was the final protection diodes got quite hot during a 62 amp load test on a single bank for 4 each 13.5 vdc supplies providing 53 vdc (after the diode drop) in series. I measured at least 150F on them possibly as high as 180F. These diodes have a max temp of 150C which equals 302F. So it wasn't catastrophically hot. But hot enough to want to try to manage it better than doing nothing. I had initially thought I would put some thermal paste under them and bolt them onto the case bottom. However, since I had some spare parts to do better than that I decided to opt for something a lot more robust. I had some spare heat sink material and I have a lot of spare computer fans around so I decided to try to mate them up and build to identical heat sink setups for each set of diodes. Each bank of 4 supplies in this (there are 2 banks) has a dual diode at the end before both banks merge into a single output. This is to keep their output separated from each bank.
So here are some photo's of what I came up with. I still have to apply the new thermal paste under them, and make some new cables to run to the output terminals, but I was able to get things migrated around in the case, and the completed heat sink / fan combos built and installed this past weekend. These fans blow INTO the heat sink instead of sucking out of them. This forces most of the air UP the fins and out of the case through exhaust fans at the top of the case.
I discovered through some recent load testing that at just a bit above 1/2 max load on the power supply the dual protection diodes I'm using get quite hot, at least 150F degrees. So to help control that I'm going to cut up a large heat sink I have tomorrow and mount these 200 amp protection diodes on them. I'm probably also going to add fans to the heat sinks as well. Here are some pictures.
Creating a DC load testing device is pretty simple should you ever need to do this. It was also very helpful to point out some short comings in my DC power supply design.
Use a cheap 10 gallon (or more) METAL barrel.
Drill 1"+ holes in the barrel to accommodate the amount of "AC" water heating elements you intend to use. (I used a step-bit to get this very tight, in order to avoid leakage)
Clean the metal around the whole at least 1/2" with a wire brush or sand paper.
Use high temp silicon like (RTV "Red" which is good to 650F, or SIL-100-GP which is good to 450F) to help seal inside and out of the hole you drilled.
Use high temp thread sealer between the water heater element and the retaining nut.
Use Stainless Steel 1" NPS nuts (probably will cost more than the heating elements!) mine where $7.50 for the nuts, and the heating elements were $5.79.
Heating elements I used are 120v/1500w @ 10 ohms. Which makes the calculations fairly simple. It's basically just a 10 ohm resistor. two in parallel = 5 ohm, three = 3.33 ohm, four = 2.5 Ohm, six = 1.6 Ohm etc... Once you know the ohms of the heating elements you're using it's just (ohm/#ele)=resulting ohm. Then using ohm's law you can calculate the amps @ volts you'll be pulling through it.
WARNING...Be sure to use a wire-gauge that can handle the amps you've calculated! Or you've basically created a fire-starter.
That's basically it. Wire the heating elements in parallel and fill the barrel with water, connect power and test. BE SURE TO Monitor the water temp as well as the heating elements and wires used.
WARNING...if you don't understand what I'm talking about, don't attempt to do this!
I discovered that at 33 amps I saw virtually no stress on any components I was using. It wasn't until I doubled that by adding 6 more heating elements to the load and drew 62 amps @ 53vdc that I noticed my final protection diodes were hot and needed to be heat-sinked and possibly a small fan added. (more to come on that in the next set of tests).
I discovered at 62 amps at TURN-ON time (instant load on the power supply when 240vac was applied) that sometimes individual 13.5vdc power supplies would startup and immediately go into 'safe-mode' and shut back down. That's a safety feature built into the dps-1200fb power supplies.
This was happening because I have no disconnect-relay or 'contactor' device at the final of the power supply. So any load attached at start up would cause an instant surge before the bank of supplies had a chance to stabilize, and they would go into safe-mode. Which basically means I would have to power it all down and try again and hope that the next time they'd start up (and they usually would).
This isn't acceptable behavior.
I also already have the ability to 'act' if a bank of supplies isn't putting out the voltage I expect. It's already built into this system, but because I have no final relay/contactor to energize I can't really 'act'. Even tho I have the monitor and relay in place to do so.
Resolution of the last two items in this list, I've found and ordered a 12vdc coil, 220 Amp @ 55vdc 'Contactor'. below are some pix of it. I found this one on Ebay "New" for about $43.00. I'll add a 10 second startup timer which will fire at power on time. After 10 seconds the power supplies should be stable enough for a good sized surge.
I can also wire this same contactor up to the bank voltage monitors so that if something fails and I don't have the voltage I expect output power will not be available because I can force this contactor to go 'open'.
The first attempt failed to start all 4 supplies in bank 0.
This afternoon I drilled and filed out the new holes for the 6 addition water heating elements being added to double the load test on the power supply from 33 amps to 62 amps. And also added both a water inlet valve and outlet value (3/4"). This will allow for water flow into and out of the tank. Cold water comes in the bottom and hot water drains out the top.
For this test I didn't run water flow through it. I simply filled the can up and ran the test since it was basically a short 'smoke test'. I ran into two small issues this time around. I was only running "bank 0" so 4 each 13.5vdc supplies in series. Instead of bank 0 and bank 1 which would be 8 supplies. The idea was to put a good strain on the system. So what I found is that the 'final' dual diode warms up pretty good. I measured about 150F on it which is too high. I need to heat sink those final diodes and possibly even add a fan, or TEC to them to really keep them cool when there's load. This test was just between 1/2 and a 3/4 load test. 75 amps would = 3/4 and this was drawing 62 amps. These diodes are rated for 150v @ 100 amps. The power supply bank that feeds them is capable of 100 amps. However I am running two wires from the positive side out which split the load one wire goes to one of the dual diodes sides the other wire to other side. I suspect this is sharing the load somewhat but I haven't tested that. I might not be too :-) Won't know for sure until I test it out next time.
I also found that the protection diodes on the power supplies where about 90F. This could just be heating up from the power supply they're connected to...and not because they're taking any sort of load (because they shouldn't be). I found that with this much load on the system instantly at start up...all of the power supplies didn't want to come up. The first attempt I had about 26v the second was 53v the third try was 40v. So what's happening is that these power supplies don't like having that heavy load placed on them instantly at start up and they'll power down (what appeared to be randomly). This didn't happen when I was pulling 33 amps. This is the first I've seen it happen, so the load test process is helping me find things I need to address. I have been thinking for a while that I need a FINAL relay that can be controlled from a few places. Like when the output voltage isn't where it should be for one of the banks I would want to OPEN the FINAL RELAY. But also to solve this 'instant load at startup' instability (which won't actually happen when I'm using this with the SSPA anyway, but I'm just trying to make this be robust and survive me being stupid in the future. So to solve that I just need to add a timer to control a final relay and tell it to wait 10 seconds while the supplies startup and stabilize, and then close the relay and draw power as needed. I've thought I needed something like that for a while, but I've been putting it off. I'm pretty sure I know the relay I want to use, but I need to find one for sale. It may take a while to find one that can be controlled by 12vdc and be able to handle 54vdc @ 200 amps :-) But I'm sure I can find one.
Today I ordered 6 more 120v / 1500 watt / 10 ohm water heater elements to add to my load-test-barrel :-) That'll bring the amperage up to about 64-66 amps or so. Should be a good solid test for this 200 amp capable power supply. The heating elements are fairly cheap at around $5.80 each. So no big deal there. But the 1" NPS Stainless Steel nutz that go inside the barrel to mount them are almost $8.00 each. (I'd try a different nut if I could find one, if I was doing this over again, or if I built another like this, but fortunately I won't have to build another one like this cause this one is all mine :-) I *MAY* purchase another 6 elements and nuts sometime to get this load tester up to 100 amp capable. But I don't think I'd go much higher. So...in about a week or so I should have these 6 I ordered today added into the load test barrel and present a test here on the blog. 66 amps should be enough to press these 8 supplies hard enough so that the fans will come on and the system will generate some heat I think. (besides the water in the water in the barrel). If not, then I'll have no more worries about this supplies ability to be pressed VERY hard. So far at 33 amps it's not even feeling it. Not at all. Things are about the same was when it's all on idle! Seriously. That's pretty impressive to me. So, more to come when the next parts come in. After this next round of tests, I'll be ready to start getting the RF amplifier parts ordered and built ready to be used with this power supply.
UPDATE:
Parts all arrived together yesterday, I'll be upgrading the Load Tester later today 9/22.
Ran the first long term test at load today. 33 amps @ 53 vdc - I was running both banks for this test.
I KNOW this is a bazaar looking way to test a DC power supply. I'm not sure I've ever seen anything like it before. But it's similar to a dummy-load for RF where there's a resistor in a gallon of oil I guess.
Except that these are 120v AC 1500 watt, 10 Ohm Heating elements (6 each) fitted into the sides of a 10 gallon barrel of water and wired all in parallel. Which results in about 1.6 Ohms of resistance @ 53 vdc.
This test started with 76 F degree water. I ran the test until the temperature went up about 10 F degrees so about 86 F.
The 8 DSP-1200fb 13.5vdc power supplies running on 240v AC only reached about 76-80F degrees, and the fans barely came of idle during the entire 6-7 minute test.
Everything ran just fine.
The only thing that didn't work as expected is the Ammeter on the face plate of the power supply controls. It was reading about 15 amps. While the Fluke Clamp meter was indicating 33 amps. So I'll have to figure out what's happening with that.
Here's a Video of this test session.
The next test I do will only running a single bank of 4 each 13.5 vdc supplies instead of the two banks of 8 total I ran today. That should stress the 4 supplies a bit harder, I think enough to run their fans off idle.
To leak-proof I used "SIL 100 GP" silicone which is good to temp's of 450 F degrees! WAY overkill, but I had a 1/4 tube of this left over from some other project. I also used high-temp thread sealer for the 1" NPS Stainless Steel nuts that go inside the tank to retain the heating elements. I used a heating element socket (they're cheap and the best way to tighten them down). And a pair of vice grips on the inside holding the nut. This was all about as simple as can be.
The hardest part of building this load tester was just cutting the 1"+ holes for the heating elements with a cheap set of 'step-bits'. And even that wasn't hard really.
I may order 12 more heating elements now that I've seen that 6 of them barely heats the water 10 degrees in 6-7 minutes. With a total of 18 heating elements I should be able to get to 100 amps being drawn. That's where I want to be ultimately for testing this power supply. 10 Ohms / 6 elements = 1.6 Ohms. So 10 Ohms / 18 elements = 0.55 Ohms. Which would be close to 100 amps being drawn.
The new 10 Gallon water tight steel barrel is here, as are my 6 each 120vac/1500w/10Ohm water heater elements and some Stainless Steel 1" NPS with o-ring seals. I bought a cheap set of 'step-drill bits' that step from 1/8" to 1 3/8".
I semi-randomly cut 6 holes in my new water tight barrel and inserted the water heater elements and attached the 1" NPS nuts and o-rings. This will end up being my high-amperage load-tester. There's plenty of space for 12 more heater elements if I wanted to pull 100 Amps @ 53vdc with this setup it would take a total of 18 water heater elements. So once that's been done, the next step is to add a INLET and OUTLET from the 'tank' (barrel) so that I can flush cold water though it while it's in use. If I didn't care that I 'wasted' water I could just just use garden hose and bulk-head fittings and run water out of the 'tap' through the tank and out into the garden or something. But what I'm going to do is locate a small car radiator, and 12vdc fan and a water DC water pump and just recirculate the water I think. My first test will probably be the garden-hose method :-) Because it's quick and simple and should help me locate any leak issues. Here's some of the construction pictures. So far the only expensive bits where the 10 Gallon Barrel (I paid $25 shipping :( ) and the SS nuts which were $45.00. I could really have just used any 1" NPS nut but these were about as perfect as I could find. 6 each heating elements costs $30.00 (amazon prime makes shipping free). So this'll test up to about 33 amps @ 53 vdc as I've already discovered. $50.00 more in heating elements and 12 more holes in my barrel will get me right up to 100 amps. Which is the max that one "bank" of 4 supplies in series will provide.
Today I connected 5 each (1500w/120v/10 Ohm) water heater elements (costing $5.00 each) in parallel which ended up with between 1.9 and 2 Ohm's of resistance. I was able to witness the voltage drop over the final Diodes from 53.9v to 53v. Which is very close to what the datasheet predicted for these dual diodes. I ran both bank(1) and bank(2) for this test today. The result was 26.5 amps pulled with a 0.9vdc drop after the final protection diodes voltage of 53.0v. Based on the how the paralleled heating elements ohm'd at 1.9 Ohms the prediction was 27.89 amps @ 53v for 1478 watts total. So prediction math vs. reality was off about 1.49 amp but at least the supply can draw nearly 30 amps using this test. I have a 6th heating element I could connect as well. That would theoretically end up at 1.6 Ohms and 33.1 amps drawn. Currently the problem I have is that I can't run this test and let things remain at-load for more than about 10 seconds. I don't want these heating elements to overheat. And after just 10 seconds they are to hot to touch. I have a 10 gallon steel drum that I'm going to tap for these six heating elements. Then I'll fill it with water. That should allow for several minutes of testing. Apparently it would take abut 167 minutes or 2 Hours and 47 Minutes to 'boil' 10 gallons of water if Efficiency = 95% and the starting water temp was 50F degrees. So I'll probably add ICE to the water and get good and cold before I start any real testing. Although if the starting water temp was 80F degrees it would still take 136 Minutes to boil or 2 Hours and 16 Minutes. The idea is to not let the water get anywhere near that hot. Testing once that's working will be one minute ON on minute off (to simulate something like JT65A load's). For reference 30 amps at 53 vdc is enough to drive a single BLF188XR to full power output.
UPDATE
I added the 6th heating element and indeed the Ohm's dropped to 1.6 Ohms when tested it yielded a draw of 32.4 amps.
My 10 Gallon Steel barrel will be here tomorrow, and I'll try to figure out how I'm gonna mount 6 heating elements into it :-) More on that another soon I hope.
