As a sort of follow-up to the previous posting about the RX-888 (Mk2) I decided to make some measurements to help characterize the gain and attenuation settings.

The RX-888 (Mk2) has two mechanisms for adjusting gain and attenuation:

• The PE4312 attenuator.  This is (more or less) right at the HF antenna input and it can be adjusted to provide up to 31.5dB of attenuation in 0.5dB steps.
  

• The AD8370 PGA.  This PGA (Programmable Gain Amplifier) can be adjusted to provide a "gain" from -11dB to about 34dB.

Note:

While this blog posting has specific numbers related to the RX-888 (Mk2), its general principles apply to ALL receivers - particularly those operating as "Direct Sampling" HF receivers.  A few examples of other receivers in this category include the KiwiSDR and Red Pitaya - to name but two.

Other article RX-888 articles:

RX-888 Thermal issues:  I recently posted another article about the RX-888 (Mk2) discussing the thermal properties of its mechanical construction - and ways to improve it to maximize reliability and durability.  You can find that article here:  Improving the thermal management of the RX-888 (Mk2) - link. 

Using an external clock with the RX-888:  The 27 MHz external clock input to the RX-888 is both fragile and fickle.  To learn a bit more about how to reliably clock an RX-888 from an external source, read THIS article.

* * * * *

Taking measurements

To ascertain the signal path properties of an RX-888 (Mk2) I set its sample rate to 64 Msps and using both the "HDSDR" and "SDR Radio" programs (under Windows - because it was convenient) and a a known-accurate signal generator (Schlumberger Si4031) I made measurements at 17 MHz which follow:

Gain setting (dB)	Noise floor (dBm/Hz)	Noise floor (dBm in 500Hz)	Apparent Clipping level (dBm)
-25	-106	-79	>+13dBm
+0	-140	-113	+3
+10	-151	-124	-8
+20	-155	-128	-18
+25	-157	-130	-23
+33	-158	-131	-31

Figure 1:  Measured performance of an RX-888 Mk2.  Gain mode is "high" with 0dB attenuation selected.

For convenience, the noise floor is shown both in "dBm/Hz" and in dBm in a 500 Hz bandwidth - which matches the scaling used in the chart below.  As the programs that I used have no direct indication of A/D converter clipping, I determined the "apparent" clipping level by noting the amplitude at which one additional dB of input power caused the sudden appearance of spurious signals.  Spot-checking indicated that the measured values at 17 and 30 MHz were within 1 dB of each other on the unit being tested.

Determining the right amount of "gain"

It should be stated at the outset that most of the available range of gain and attenuation provided by the RX-888's PE4312 step attenuator and AD8370 variable gain amplifier are completely useless to us.  To illustrate this point, let's consider a few examples.

Consider the chart below:

Figure 2:  ITU chart showing various noise environments versus frequency.

This chart - from the ITU - shows predicted noise floor levels - in a 500 Hz bandwidth - that may be expected at different frequencies in different locations.  Anecdotally, it is likely that in these days of proliferating switch-mode power supplies that we really need another line drawn above the top "Residential" curve, but let's be a bit optimistic and presume that it still holds true these days.

Let us consider the first entry in Figure 1 showing the gain setting of 0dB.  If we look at the "Residential" chart, above, we see that the curve at 30 MHz indicates a value very close to the -113dBm value in the "dBm in 500 Hz" column.  This tells us several things:

• Marginal sensitivity.  Because the noise floor of the RX-888 (Mk2) and that of our hypothetical RF environment are very close to each other, we may not be able to "hear" our noise floor at 30 MHz (e.g. the 10 meter amateur band).  One would need to do an "antenna versus no antenna" check of the S-meter/receiver to determine if the former causes an increase in signal level:  If not, additional gain may be needed to be able to hear signals that are at the noise floor.
  

• More gain may not help.  If we do perform the "antenna versus no antenna" test and see that with the antenna connected we get, say, an extra S-unit (6dB) of noise, we can conclude that under those conditions that more gain will not help in absolute system sensitivity.
  

Thinking about the above two statements a bit more, we can infer several important points about operating this or any receiver in a given receive environment:

• If we can already "hear" the noise floor, more gain won't help.  In this situation, adding more gain would be akin to listening to a weak and noisy signal and expecting that increasing the volume would cause the signal to get louder - but not the noise.  
  

• More gain than necessary will reduce the ability of the receiver to handle strong signals.  The HF environment is prone to wild fluctuations and signals can go between well below the local noise floor and very strong, so having any more gain that you need to hear your local noise floor is simply wasteful of the receiver's signal handling capability.  This fact is arguably more important with wide-band, direct-sampling receivers where the entire HF spectrum impinges on the analog-to-digital converter rather than a narrow section of a specific amateur band as is the case in "conventional" analog receivers.

Let us now consider what might happen if we were to place the same receiver in an ideal, quiet location - in this case, let's look at the "quiet rural" (bottom line) on the chart in Figure 2.

Again looking at the value at 30 MHz, we see that our line is now at about -133dBm (in 500 Hz) - but if we have our RX-888 gain set at 0 dB, we are now ((-133) - (-113) = ) 20 dB below the noise floor.  What this means is that a weak signal - just at the noise floor - is more than 3 S-units below the receiver sensitivity.  This also means that a receiver that may have been considered to be "Okay" in a noisy, urban environment will be quite "deaf" if it is relocated to a quiet one.

In this case we might think that we would simply increase our gain from 0 dB to +33dB - but you'll notice that even at that setting, the sensitivity will be only -131dBm in 500 Hz - still a few dB short of being able to hear the noise in our "antenna versus no antenna" test.

Too much gain is worse than too little!

At this point I refer to the far-right column in Figure 1 that shows the clipping level:  With a gain setting of +33dBm, we see that the RX-888 (Mk2) will overload at a signal level of around -31dBm - which translates to a  signal with a strength a bit higher than "S9 + 40dB".  While this sound like a strong signal, remember that this signal level is the cumulative TOTAL of ALL signals that enter the antenna port.  Thinking of it another way, this is the same as ten "S9+30dB" signals or one hundred "S9+20dB" signals - and when the bands are "open," there will be many times when this "-31dBm" signal level is exceeded from strong shortwave broadcast signals and lightning static.

In the case of too-little gain, only the weakest signals, below the receiver's noise floor will be affected - but if the A/D converter in the receiver is overloaded, ALL signals - weak or strong - are potentially disrupted as the converter no longer provides a faithful representation of the applied signal.  When the overload source is one or more strong transmissions, a melange of all signals present is smeared throughout the receive spectrum consisting of many mixing products, but if the overload is a static crash, the entire receive spectrum can be blanked out in a burst of noise - even at frequencies well removed from the original source of static.

Most of the adjustment range is useless!

Looking carefully at Figure 1 at the "noise floor" columns, you may notice something else:  Going from a gain of 0 dB to 10 dB, the noise floor "improves" (is lower) by about the same amount - but if you go from 25 dB gain to 33 dB gain we see that our noise floor improves by only 1 dB - but our overload threshold changes by the same eight dB as our gain increase.

What we can determine from this is that for practical purposes, any gain setting above 20 dB will result in a very little receiver sensitivity improvement while causing a dramatic reducing in the ability of the receiver to handle strong signals.

Based on our earlier analysis in a noise "Urban" environment, we can also determine that a gain setting lower than 0 dB will also make our receiver too-insensitive to hear the weakest signals:  The gain setting of -25dB shown in Figure 1 with a receive noise floor of -79dBm (500 Hz) - which is about S8 - is an extreme example of this.

Up to this point we have not paid any attention to the PE4312 attenuator as all measurements were taken with this set to minimum.  The reason for this is quite simple:  The noise figure (which translates to the absolute sensitivity of a receiver system) is determined by the noise generation of all of the components.  As reason dictates, if you have some gain in the signal path, the noise contribution of the devices after the gain have lesser effects - but any loss or noise contribution prior to the gain will directly increase the noise figure.

Note:

For examples of typical HF noise figure values, see the following articles:

• Measurements on a Multiband R2P Pro Low-Noise Amplifier System - link to the Web Archive.  Refer to page 9 section 7.
• Antenna and Receiver Noise Figure - link.
  

Based on the articles referenced above, having a receiver system with a noise figure of around 15dB is the maximum that will likely permit reception at the noise floor of a quiet 10 meter location.  If you aren't familiar with the effects of noise figure - and loss - in a receive signal path, it's worth playing with a tool like the Pasternack Enterprises Cascaded Noise Figure Calculator (link) to get a "feel" of the effects.

I do not have the ability to measure the precise noise figure of the RX-888 (Mk2) - and if I did do so, I would have to make such a measurement using the same variety of configurations depicted in Figure 1 - but we can know some parameters about the worst-case:

• Bias-Tee:  Estimated insertion loss of 1dB
  
• PE4312:  Insertion loss of 1.5dB at minimum attenuation
• RF Switch (HF/VHF):  1dB loss
  
• 50-200 Ohm transformer:  1dB loss
  
• AD8370 Noise figure:  8dB (at gain of 20dB)

The above sets the minimum HF floor noise figure of the RX-888 (Mk2) at about 12.5dB with an AD8370 gain setting of 20dB - but this does not include the noise figure of the A/D converter itself - which would be difficult to measure using conventional means.

On important aspect about system noise figure is that once you have loss in a system, you cannot recover sensitivity - no matter how much gain or how quiet your amplifier may be!  For example, if you have a "perfect" 20 dB gain amplifier with zero noise, if you place a 10 dB attenuator in front of it, you have just turned it into an amplifier with 10 dB noise figure with 10dB gain and there is nothing that can be done to improve it - other than get rid of the loss in front of the amplifier.

Similarly, if we take the same "perfect" amplifier - with 20dB of gain - and then cascade it with a receiver with a 20dB noise figure, the calculator linked above tells us that we now have a system noise figure of 3 dB since even with 20dB preceeding it, our receiver still contributes noise!

If we presume that the LTC2208 A/D converter in the RX-888 has a noise figure of 40dB and no gain (a "ballpark" value assuming an LSB of 10 microvolts - a value that probably doesn't reflect reality) our receive system will therefore have a noise figure of about 22dB.

What this means is that in most of the ways that matter, the PE4312 attenuator is not really very useful when the RX-888 (Mk2) is being used for reception of signal across the HF spectrum, in a relatively quiet location on an antenna system with no additional gain.

Where is the attenuator useful?

From the above, you might be asking under what conditions would the built-in PE4312 attenuator actually be useful?  There are two instances where this may be the case - and this would be applied ONLY if you have been unable to resolve overload situations by setting the gain of the AD8370 lower.

• In a receive signal path with a LOT of amplification.  If your receive signal path has - say - 30dB of amplification (and if it does, you might ask yourself "why?") a moderate amount of attenuation might be helpful.
  

• In a situation where there are some extremely strong signals present.  If you are near a shortwave or mediumwave (AM broadcast) transmitter that induces extremely strong signals in the receiver that cause intractable overload, the temporary use of attenuation may prevent the receiver from becoming overloaded to the point of being useless - but such attenuation will likely cause the complete loss of weaker signals.  In such a situation, the use of directional antennas and/or frequency-specific filtering should be strongly considered!
  

Improving sensitivity

Returning to an earlier example - our "Quiet Rural" receive site - we observed that even with the gain setting of the RX-888 (Mk2) at maximum, we would still not be able to hear our local noise floor at 30 MHz - so what can be done about this?

Let us build on what we have already determined:

• While sensitivities is slightly improved with higher gain values, setting the gain above 20dB offers little benefit while increasing the likelihood of overload.
  

• In a "Quiet Rural" situation, our 30 MHz noise floor is about -133dBm (500 Hz BW) which means that our receiver needs to attain a lower noise floor than this:  Let's presume that -136dBm (a value that is likely marginal) is a reasonable compromise.

With a "gain" setting of 20dB we know that our noise floor will be around -128dBm (500 Hz) and we need to improve this by about 8 dB.  For straw-man purposes, let's presume that the RX-888 (Mk2) at a gain setting of 20dB has a noise figure of 25dB, so let's see what it takes for an amplifier that precedes the RX-888 (Mk2) to lower than to 17dB or so using the Pasternak calculator above:

• 10dB LNA with 7 dB noise figure:  This would result in a system noise figure of about 16 dB - which should do the trick.

Again, the above presumes that there is NO  loss (cable, splitters, filtering) preceding the preamplifier.  Again, the presumed noise figure of 25dB for the RX-888 (Mk2) at a gain setting of 20 is a bit of a "SWAG"  - but it illustrates the issue.

Adding a low-noise external amplifier also has another side-effect:  By itself, with a gain setting of +33, the RX-888 (Mk2)'s overload point is -31dBm, but if we reduce the gain of the RX-888 to 20dB the overload drops to -18dBm - but adding the external 10dB gain amplifier will effectively reduce the overload to -28dBm, but this is still 5 dB better than if we had turned the RX-888's gain all of the way up!

Taking this a bit further, let's presume that we use, instead, an amplifier with 3dB noise figure and 8 dB gain:  Our system noise figure is now about 17dB, but our overload point is now -26dBm - even better!

The RX-888 is connected to a (noisy) computer!

Adding appropriate amounts of external gain has an additional effect:  The RX-888 (and all other SDRs) are computer/network connected devices with the potential of ingress of stray signals from connected devices (computers, network switches, power supplies, etc.).  The use of external amplifiers can help override (and submerge) such signals and if proper care is taken to choose the amount of gain of the external amplification and properly choose gain/attenuation settings within the receiver, superior performance in terms of sensitivity and signal-handling capability can be the result.

Additional filtering

Only mentioned in passing, running a wideband, direct-sampling receiver of ANY type (be it RX-888, KiwiSDR, Red Pitaya, etc.) connected to an antenna is asking a lot of even 16 bits of conversion!  If you happen to be in a rather noisy, urban location, the situation is a bit better in the sense that you can reduce receiver gain and still hear "everything there is to hear" - but if you have a very quiet location that requires extra gain, the same, strong signals that you were hearing in the noisy environment are just as strong in the quiet environment.

Here are a few suggestions for maximizing performance under the widest variety of situations:

• Add filtering for ranges that you do not plan to cover.  In most cases, AM band (mediumwave) coverage is not needed and may be filtered out.  Similarly, it is prudent to remove signals above that in which you are interested.  For the RX-888 (Mk2), if you run its sampling rate at just 65 MHz or so, you should install a 30 MHz low-pass filter to keep VHF and FM broadcast signals out.
  

• Add "window" filtering for bands of interest.  If you are interested only in amateur radio bands, there are a lot of very strong signals outside the bands of interest that will contribute to overload of the A/D converter.  It is possible to construct a set of filters that will pass only the bands of interest - but this does not (yet?) seem to be a commercial product.  (Such a product may be available in the near future - keep a lookout here for updates.)
  

At the Northern Utah WebSDR (link) we run a number of KiwiSDR receivers.  These receivers, which are inherently broadband (10 kHz to 30 MHz) allow a limited number of users to tune across the bands, allowing reception on frequencies that are not covered by the WebSDR servers.

At present there are six of these receivers on site:  Three are connected to the TCI-530 Omnidirectional antenna (covering 630-10 meters - 2200 meters is included via a separate E-field whip), two are on the east-pointing log-periodic beam antenna (which overs 40-10 meters) and the newest is connected to the northwest-pointing log-periodic beam antenna (which covers 30-10 meters).

Figure 1: Power supply in a PC case! The PC case housing the power supply was repurposed - because, why not? Click for larger version

The power requirements of a KiwiSDR are modest, being on the order of 600-800 mA, but the start-up current can briefly exceed 1.25 amps.  Additionally, they do not start up reliably if the voltage "ramps up" rather slowly - a problem often exacerbated by the fact that the extra current that they draw upon power-up can cause a power supply to "brown out".

Up to this point we had been running 5 KiwiSDRs:  Three of them were powered by a pair of 5 volt, 3 amp linear power supplies that are "dioded-ANDed" together to form a 6 amp power supply and the other two KiwiSDRs were powered from a heavily-filtered 5-volt, 3 amp switching power supply.

In recent months, the dual 3 amp linear supply had become problematic, not being able to handle the load of the three KiwiSDRs, so we had to power down KiwiSDR #3.  With the recent installation of the northwest-pointing log periodic antenna, we were also looking toward installing another KiwiSDR for that antenna and we were clearly out of power supply capacity.

Using an ATX supply as a general-purpose power supply - it's not just the green wire!

If you look around on the Web, you'll see suggestions that you just "ground the green wire" to turn on an ATX supply, at which point you may use it as a general-purpose supply.  While grounding the green wire does turn it on, it's not as simple as that - particularly if you leave the power supply unattended.

For example, what if there is a brief short on the output while you are connecting things, or what if the power browns out (or turns off) for just the "wrong" amount of time.  These sorts of things do happen, and can "trip out" the power supply and it may never restart on its own.

With the site being remote, we couldn't afford for this to happen - so you'll see, below, how we remedied this.

Putting together another power supply:

With six KiwiSDRs, the power supply requirements were thus:

• 5 amps continuous, making the assumption that a KiwiSDR's average current consumption would be about 830 mA - a number with generous overhead.
  

• 9 amps on start-up, presuming that each KiwiSDR would briefly consume 1.5 amps upon power-up, again a value with a bit of overhead.
  

• The power supply must not exhibit a "slow" ramp-up voltage as the KiwiSDRs did not "like" that.

In looking around for a power supply on which to base the design, the obvious choice was an computer-type ATX power supply.  Fortunately, I have on-hand a large number of 240 watt ATX supplies with active power factor correction which are more than capable of supplying the current demands, being rated for up to 22 amps load on the 5 volt supply - more than enough headroom as I would be needing less than half of that, at least with the currently-planned usage.

Circuit description:

Refer to the schematic in Figure 2 for components in the description.

Added filtering:

While these power supplies were already known to be adequately RF-clean (important for a receive site!) from their wide use for the WebSDR servers because we would be conducting the DC outputs outside the box - and to receivers - I felt it important that additional filtering be added.  Having scrapped a number of PC power supplies in the past, I rummaged around in my box of random toroids and found two that had probably come from old PC power supplies, wound with heavy wire consisting of 4 or 5 strands in parallel.  These inductors measured in the area 10s of microHenries, enough for HF filtering when used with additional outboard capacitance.

These filter networks were constructed using old-fashioned phenolic terminal lug strips.  These consist of a row of lugs to which components are soldered - typically with one or two of the lugs used for mounting, and also "grounding".  Rather than mount these lugs using a drill and screw, they were soldered to the steel case itself - something easily done by first sanding a "bare" spot on the case to remove any paint or oxide and then using an acid-core flux - cleaning it up afterwards, of course!

The heavier components (inductors, capacitors) were mechanically secured using RTV (silicone) adhesive to keep them from moving around - and to prevent the possibility of the inductor's wire from touching the case and chafing.

Looking at the schematic you may note that  C202, C302, C501, C502 and C503 are connected to a "different" ground than everything else.  While - at least for this power supply - the "Common" (black) wire is internally connected to the case, it's initially assumed that this lead - which comes from the power supply - may be a bit "noisy" in terms of RF energy, so they are RF bypassed to the case of the power supply.  This may have been an unneeded precaution, but it was done nonetheless.

Connectorizing and wiring the power supply:

The ATX power connector was extracted from a defunct PC motherboard to allow the power supply itself to be replaced in the future if needed.  On this connector, all of the pins corresponding with the 5 volt (red wires), 12 volt (yellow wires) and ground (black wires) were bonded together to form three individual busses and heavy (12 AWG) wires were attached to each:  This was done to put as many of the wires emerging from the power supply in parallel with each other to minimize resistive losses. 

The green wire (the "power" switch) and purple wire (the 5 volt "standby") were brought out separately as they would be used as well - and the remainder of the pins (3.3 volt, -12 volt, -5 volt, "power good", etc.) were flooded with "hot melt" glue to prevent anything from touching anything else that it shouldn't.

The 5 volt supply was split two ways - each going to its own L/C filter network (L501, L502, C502, C503, C504, C505) as shown in the schematic, this being done to reduce the total current through the inductor - both to minimize resistive losses, but also to reduce the magnetic flux in each inductor, something that could reduce its effective inductance.

Although I don't have immediate plans to use the 12 volt supply, a similar filter (L503, C506, C507) was constructed for the 12 volt supply lead.  On the output side of the 12 volt filter, a 3 amp self-resetting thermal fuse (F501) was installed to help limit the current should a fault occur. 

About the self-resetting fuses:

 These fuses - which physically look like capacitors - operate by having a very low resistance when "cold".  When excess current flows, they start to get warm - and if too much current flows, they get quite hot (somewhere above 200F, 100C) and their internal resistance skyrockets, dropping the current to a fraction of its original value:  It's this current flow and their heat that keeps the resistance high.

It's worth noting that these fuses don't "disconnect" the load - they just reduce the current considerably to protect whatever it is connected to it.  Since, when "blown", they are hot, they must be mounted "in the clear" away from nearby objects that could be damaged by the heat - and also to prevent lowering of their trip current by trapping heat or being warmed by another component - such as another such fuse.  

It should be noted that if the outputs - either 5 or 12 volts - are "hard shorted", the thermal fuse may not react quickly enough prevent the power supply from detecting an overcurrent condition and shutting down.  As an output short is not expected to be a "normal" occurrence, this behavior is acceptable - but it will require that the power supply be restarted to recover from shutdown, as described below.

In the case of the KiwiSDRs, they are connected with fairly long leads (about 6 feet, 2 meters) and often have enough internal resistance to reduce the current below the power supply's overcurrent limit and rather than allowing the full current of the power supply (which could be more than 20 amps) to flow through and burn up this cable, the fuse will trip as it should, protecting the circuit.  To "reset" the fuse, the current must be removed completely for long enough for the device to cool - something that is done with the 5 volt supplies as we'll see, below.

The controller:

As mentioned earlier, if you look on the web, you'll see other power supply projects that use an ATX power supply as a benchtop power source and most of those suggest that one simply connect the green (power on) wire to ground to turn it on - but this isn't the whole story.  In testing the power supply, I noticed two conditions in which doing this wouldn't be enough:

• Shorting a power supply output.  If the output of a good-quality ATX power supply is shorted, it will immediately shut down - and stay that way until the mains power is removed (for a minute or so) or the power supply is "shut off" by un-grounding the green wire for a few seconds before reconnecting to "restart" the power supply.
  

• Erratic mains power interruption.  It was also observed that if the mains power was removed for just the right amount of time, the power supply would also shut down and would not restart on its own.  It took the same efforts as recovering from an output short to restart the power supply.

Since this power supply would be at the WebSDR site - an unmanned location in rural, northern Utah - it would require additional circuitry to make this power supply usable.

Fortunately, an ATX power supply has a second built-in power supply that is independent of the main one - the "standby" power supply.  This is a low-power 5 volt supply that is unaffected by what happens to the main supply (e.g. not controlled by the power switch and not affected if it "trips off") and can be used to power a simple microcontroller-based board that can monitor and sequence the start-up of the main power supply.  For this task I chose the PIC16F688, a 14 pin microcontroller with A/D conversion capability and a built-in clock oscillator.

As seen in the schematic, the "5 volt standby" is dioded-ORed (D601, D602) with the main power supply (12 volts) so that it always gets power - from either the 5 volt standby, or from the 12 volt output - when mains is applied.  R603 and capacitor C602 provide a degree of protection to the voltage regulator should some sort of "glitch" appear on the 12 volt supply - possibly due to the 5 volt load being abruptly disconnected (or connected) as the 5 and 12 volt supplies are "co-regulated" in the sense that it's really only the 5 volt output that is being regulated well - the 12 volt power supply's output is pretty much a fixed ratio to the 5 volt and doesn't really have much in terms of separate regulation.

It should be noted that when operating from the standby +5 volt power source, the voltage from U2 (the 5 volt regulator) is on the order of 3 volts or so (drop through D602 and U2) but this is comfortably above the "brownout" threshold of the PIC, which is around 2.5 volts, so there isn't really a worry that the low-voltage brownout detector will trigger erroneously and prevent start-up.  If it had, I would have simply moved the cathode side of D602 to the +5V side of U2.

Figure 3:  Inside the case! Top right:  12 volt supply filtering and thermal fuse Upper-middle:  Dual 5 volt filtering Lower middle:  Controller board with FET switches and thermal fusing. The ATX power supply is in the lower-left corner. Click on the image for a larger version.

Because the PIC microcontroller can monitor the 12 volt supply (via R601/R602) it "knows" when the main ATX supply is turned off.  Through the use of an NPN transistor (Q401) - the collector of which can be used to "ground" the green "power on" line, the controller can turn the main power supply on and off as follows:

• When the microcontroller starts up, it makes sure that the ATX "power on" wire is turned off (e.g. un-grounded).  This is done by the microcontroller turning off Q401.
  

• After a 10 second delay, it turns on the power supply by turning on Q401.
  

It also monitors the power supply to look for a fault.  If either the 5 or 12 volt output is shorted or faults out, both power supply outputs (but not the 5 volt "standby" output) disappear.

• If, while running, the monitored 12 volt supply (via R601/R602 and "12V V_MON") drops below about half the voltage (e.g. trips out) the "power on" wire is turned off using Q401, disabling the ATX power supply.
  

• A 10 second delay is imposed before attempting to turn the power supply back on.
  

• Once the power supply is turned back on, monitoring of the voltage resumes.

In practice, if there is a "hard" short on the output, the power supply will attempt to restart every 10 seconds or so, but remember that a short on an output could occur with ANY sort of power supply, so this isn't a unique condition.

5 volt output sequencing and monitoring:

The other function of the controller is to sequence and monitor the 5 volt outputs.  As mentioned earlier, it was noted that the KiwiSDRs do not "like" a slow voltage ramp-up so a FET switch is employed to effect a rapid turn-on - and since there are two separately-filtered 5 volt busses, there are two such switches.  In order to reduce the peak current caused when the load is suddenly connected, each of these busses is turned on separately, a 10 second delay between the two of them.

The N-channel FET switches (Q203, Q303) are controlled by an NPN (Q201, Q301) transistor being turned on by the microcontroller which, in turns, "pulls" the base of a PNP transistor (Q202, Q302) low via a base resistor (R202/R302), turning it on - and other resistors (R203, R303) assure that these transistors are turned off as needed.

With the emitter of the PNP connected to the 12 volt supply, the gate voltage of the FET is approximately 7 volts higher than the drain voltage, assuring that it is turned on with adequately low resistance.  Capacitors (C201, C301) are connected between the FET's gates and sources to suppress any ringing that might occur when the power is turned on/off and as a degree of protection against source-gate voltage spikes while the 47k resistor (R207/R307) assure that the FET gets turned off.

The use of P-channel FETs was considered, but unless special "logic level" threshold devices were used, having only 5 volts between the gate and drain wouldn't have turned them fully "on" unless the -5 or -12 volt supply from the power supply was also used.  While this would certainly have been practical, N-channel FETs are more commonly available.

Figure 2:  Schematic of the ATX controller with power supply filtering, voltage monitoring, and control. See the text for a description. Click on the image for a larger version.

In series with the 5 volt supply and the FET's source is a 5 amp self-resetting thermal fuse to limit current.  Should an overload (more than 5-ish amps) occur on the output bus, this fuse will heat up and go to high resistance, causing the output voltage to drop.  If this occurs, the microcontroller, which is using its A/D converter to look at the voltage divider on the outputs (R205/R206 for the "A" channel, R305/R306 for the "B" channel) will detect this dip in voltage and immediately turn off the associated FET.  After a wait of at least 10 seconds - for the fault to be cleared (in the event that it is momentary) and to allow the thermal fuse to cool off and reset - the power will be reconnected.  If there continues to be a fault, the reset time is lengthened (up to about 100 seconds) between restart attempts.

Finally, the status of the power supply is indicated by a 2-lead dual-color (red/green) LED (LED701) mounted to be visible from the front panel.  During power supply start-up, it flashes red, during the time delay to turn on the power supplies it is yellow, when operation is normal it is green - and if there is a fault, it is red.  Optionally, another LED (LED702) can be mounted to be visible:  This LED is driven with the algorithm that causes it to "breath" (fade on and off - and on, and off...) to indicate that "something" was working.  I simply ran out of time, so I didn't install it.

* * *

This power supply was put together fairly quickly, so I didn't take as many pictures as I usually would - and I omitted taking pictures of the back panel where the power supply connections are made.  Perhaps it's just as well as while I used a good-quality screw-type barrier strip, it was mounted to a small piece of 1/4" (6mm) thick plywood that was epoxied into the rectangular hole where one would connect peripherals to the motherboard.

As you would expect, the terminals are color-coded (using "Sharpies" on the wood!) and appropriately labeled.  While not pretty, it's functional!

(Comment:  The photo in Figure 3 was taken before I added the circuit to control the "Power On" wire (e.g. Q401) and the diode-OR power (D601, D602) - and it shows the dual-color LED on the board during testing.)

If you are interested in the PIC's code, drop me a note.

This page stolen from ka7oei.blogspot.com

[END]

A few weeks ago I was on vacation in remote Eastern Utah - in Canyonlands National Park, to be precise and because we had some "down time" in the evenings, after hiking, after sunset, I was able to set up a portable HF station.  Using the homebrew end-fed halfwave antenna (EFHW) of Mike, K7DOU - one end of the rope tied around a rock laying on a shelf of slick rock some 40 feet above ground level and the other end tied to a bamboo pole attached to my Jeep - I connected my FT-100 through a manual tuner as the VSWR of the EFHW wasn't necessarily very low on some of the higher bands.

Figure 1: 150 Watt Samlex sine wave inverter, sitting on the workbench. Click on the image for a larger version.

For whatever reason, I had brought along my old lap top and sound-card interface so I could work some digital modes, specifically FT-8 - a mode that I was familiar with, but had personally never worked.  The battery in my laptop had discharged, so I needed an alternate source of power and I connected my 150 watt Samlex Sine Wave inverter (a PST-15S-12A) to the battery to power the computer's power supply.

The (expected!) result of this was a tremendous "hash" all across the HF spectrum - an obvious result of the various high-power converters contained within the inverter.  On some bands the interference wasn't too bad, but on others the result was unusable.  While the battery charged, I operated on the band (20 meters, IIRC) that wasn't as badly affected.

I left the inverter running and the laptop battery charging during the cooking and eating of dinner, and with a reasonable amount of power banked I could turn off the inverter and get a zero noise floor while operating.

Why so noisy?

Modern AC inverters first convert the DC input power to something around the peak voltage found on the AC output - typically around 155 volts for 120 volt mains.  This conversion is done using a switch-mode inverter with a transformer, typically operating in the 20-60 kHz range and this output is rather rich in harmonics.

For the less-expensive "modified sine wave" inverters, the DC output is chopped, typically using an "H" bridge switch using FETs (Field Effect Transistors) with the duty cycle being varied to provide the equivalent of a 120 volt sine wave - and this switching can also add a bit of extra RFI, most notably in the form of a "buzz" - but this action produces less energy at radio frequencies than the initial voltage conversion.

The "Sine Wave" inverters perform the same step of producing the high DC voltage, but will chop the output into much smaller bits.  The method that this is done can vary, but it's sometimes done by using a "buck" type switching converter to transform the higher voltage into a varying - usually lower - voltage to simulate a sine wave on the output.  This second conversion adds yet another source of RF interference atop what is likely already the significant source that already present in the high voltage converter.

Comment:  The power converter (wall wart) that I was using to charge my laptop is particularly quiet, so I did verify that the vast majority of noise was, in fact, from the AC inverter.

Figure 2: Various mains filtering components:  All of these are bifilar, common-mode chokes, except for that in the upper-left with is a combination filter and IEC power connector. Click on the image for a larger version.

Quieting the inverter:

Fortunately, the internal space of this inverter wasn't terribly cramped so there was just enough room to add the necessary components to suppress the RF "hash" that was being conveyed on both the DC and AC lines.  While the methods of doing this sort of RF quieting have been discussed in previous blog posts (see the references at the end of this article) I'll review them in detail here.

Snap-on chokes won't do!

It's worth noting (several times!) that simply winding the power cord (DC and/or AC) around a ferrite device (e.g. a clamp-on or even a large toroid) would likely NOT be enough to solve this problem.  While doing so may knock down RFI by, perhaps, 6-10 dB - maybe 20 dB if one is really lucky - this sort of noise egress must often be attenuated by several 10s of dB to effectively quash it.  In other words, knocking down the "grunge" by 1-2 S-units is nice enough, but there will still be a lot of hash left over to bury the weakest signals! 

Internally, this inverter did pass through some rather large ferrite cylinders the DC input and (separately) AC output connections, but this very small amount of inductance would have practically no effect at all at HF - likely having been added to make a dent in the noise at VHF so that it would pass muster when subjected to EMC compliance tests.

Filtering the AC output:

I presumed (but didn't actually measure) that the majority of the noise being radiated would be from the AC output as it is "closest" to the circuits most likely to generate a lot of noise, so I concentrated most of my effort there.

The most helpful component in filtering the mains voltage output is the bifilar choke - several varieties of these being displayed in Figure 2.  This component consists of two windings in parallel on the same ferrite core - typically both leads of the mains voltage.  For the low-frequency AC currents, the halves of the choke carry equal and opposite current so there is no DC component to magnetize the core and reduce its efficacy due to saturation, but because RF energy is likely not flowing in a differential manner as is the AC mains voltage, the inductance of the two parallel windings come into effect - the magnitude of this typically being in the 10s of microHenries to milliHenries range.

Where does one get these things?  They can be found at surplus outlets if you look around, but perhaps the easiest source is from defunct PC power supplies:  These devices, found in supplies made by reputable manufacturers, are typically the first things through which the AC mains voltage pass (after any fusing) before going to the rest of the circuitry.

Figure 3: Schematic of the output filter.  While it's likely that just one bifilar inductor would have sufficed, I decided that since there was room to do so, a second one would be added for even more filtering of the "grunge" that can emanate from such a noisy circuit. Click on the image for a larger version.

 

This much inductance has significant impedance to RF energy - but inductance alone will have only limited efficacy and intrinsic capacitance of the windings will also reduce the amount of attenuation that would otherwise happen - as would have winding the mains cord/cable on a ferrite toroidal core as noted previously - so capacitors are also required to be placed strategically to help shunt away some of the residue.

Figure 4: The AC output filter in the process of being installed.  L1 and C1-C4 are mounted to the outlet itself while the connection to L2 is made using the orange leads. Click on the image for a larger version.

The diagram in Figure 3 shows the as-installed filter.  As can be seen, two separate bifilar filters (both of them being the sort as seen as the second from the lower-right in Figure 2) were used to maximize attenuation.  In this circuit, C3 and C4 are used to force any RF on the two wires to be common-mode to maximize the efficacy of the bifilar chokes' attenuation and any residual RF - which will be at rather low level and high impedance - will then be shunted to the metal case of the inverter by capacitors C1 and C2.

Figure 4 shows the installation of the filtering components in the inverter.  C1 and C2 are the disk-shaped blue capacitors seen in the upper-left, mounted directly to the inverter's single AC outlet and capacitor C3 is just in "front" of the two round disks, also mounted directly to the socket.  The first inductor, L1, can be seen in the shadows, connected to the outlet with very short, flexible leads to the plug.

Earlier, I had removed this outlet from the body of the inverter and mounted C1, C2, C3 and L1 to it and with a bit of "tetris" action, was able to reinstall the outlet back in place with the components attached.  From that point I installed C4 (to the "other" side of L1) and the (orange) connecting wires from C4 to L2, which is shown floating in space.

You might ask why there isn't another capacitor (like C4) across the "inverter" side of L2 - or other capacitors to ground other than C1/C2:  There is already a degree of filtering on the AC output of the inverter, so there is little point in adding another capacitor like C4.  As for other capacitors to "ground" like C1/C2 elsewhere in the circuitry:  These were deemed unnecessary - and doing so, particularly at the "inverter" side of L4 would simply put relatively strong RF currents onto the ground lead (e.g. inverter's case) - and our cause won't be helped in making RF currents appear where don't need them to be.  

Figure 5: Noise filter on the DC input.  It looks suspiciously like the filter on the AC output - because it's the same type, although the current-carrying capacity of L1 is much higher and the values of the capacitors are orders of magnitude larger. Click on the image for a larger version.

Filtering the DC input:

While I would presume that most of the noise would be emitted via the AC output port, filtering the DC port must be considered as well.  With the inverter's rating being 150 watts, the maximum current on the AC output would be around 1.25 amps and rather light-gauge wire could be used in the inductors - but because this same power level represents 12.5 amps at 12 volts (likely more if the battery voltage is on the low side) the filtering inductance must be made using much larger wire.

Rummaging around in my box of toroids, I found a ferrite device that was about 1" (2.54cm) in outside diameter and wound as many turns of 14 AWG flexible wire onto it as would fit (about 6 bifilar turns) and measured it to have about 30 uH of inductance per winding.  This may not seem like much, but at 1 MHz, this represents about 180 ohms of reactance.   

In referring to Figure 5, above, you'll notice that it is pretty much identical to that of the output filter - except that there is only one section of filtering.  The capacitor values are different, too:  C1 and C2 are 0.1uF units that shunt residual RF getting through L1 to ground (the case) while C3 is a low-ESR electrolytic connected across the DC leads to help force any residual AC noise on the DC lead to common-mode.  Compared to the 180 ohms of reactance of the DC bifilar choke (at 1 MHz) a good-quality, monolithic ceramic capacitor like the 0.1uF units are likely to have well under an ohm of impedance and very little of the RF hash will remain after they do their job to bypass it to the chassis ground.

Figure 6: The DC input filter.  The capacitors (not visible) are mounted to the bottom side of the terminal strip, which serves as the RF "grounding" point to the case.  L1 is just visible. Click on the image for a larger version.

Because of the limited amount of room, only one inductor was used - although it would likely be possible to have crammed another in the limited space should the above filter have proved to be inadequate (it wasn't).

As can be seen in Figure 6, a small terminal strip is visible and to it is mounted C1-C3 (not visible as they are obscured by the strip itself).  The mounting point for this strip is the ground lug near the DC input cable and the center lug is the common point for C1 and C2.

An important point to mention is the fact that this inverter - like many - have their DC and AC lines isolated from the case - and that's also important here:  Because the DC has no connection to the inverter's metal case, ALL of the DC current passes through L1 of Figure 5 - but with both halves carrying the same current, the core is not magnetized:  Magnetizing the core would likely cause it to saturate and the result would be its effective inductance plummeting - possibly reducing its efficacy as an RF filter.  It is for this reason that a bifilar choke was used on the DC input as well.

As with the AC output, the "inverter" side of L1 of Figure 5 also lacks a common-mode capacitor, but this is well represented on the input of the inverter itself with its own, built-in capacitor.

Figure 7: The final arrangement of the added filtering components.  Liberal use of RTV (silicone adhesive) was used to stabilize the components as it works well, and can be removed should repairs/modifications be required.  On the left, a generous blob of RTV has been used to keep the terminal strip's lugs at the DC input from touching the inverter's bottom cover. Click on the image for a larger version.

Additional comments:

Figure 7 shows the final arrangement of the added components.  In the upper-left corner can be seen the components of the DC input filter with come clear RTV (silicone adhesive) added to the top of the terminal strip to insulate it and keep any metal parts of it from touching the bottom cover when it was reinstalled.

On the right side is the AC output filter and on the foreground can be seen L2, now with the "hot" terminals covered by heat-shrink tubing.  This choke was first attached "temporarily" to the inverter's end plate using instant (cyanoacrylate) glue - and then several large blobs of RTV were later added to permanently hold it in place.  Just above it can be seen the orange wires that connect L2 to L1 and these components were also stabilized with rather large blobs of RTV to keep them from "flapping in the breeze".  It's worth noticing that the original ferrite cylinder is still on the AC output connection (on the black and white wires) where it connects to L4 - mainly because there was still room for it, and its efficacy, such as it is, is likely only enhanced by the addition of the new filtering components. 

Did it work?

You might ask the question:  Did this filtering work?

The answer is yes.  Placing a portable shortwave radio next to either the DC or AC power leads from the inverter, one can't detect that it is running at all.  If the radio is placed right atop the inverter, some hash can be detected, but this is likely from direct radiation of magnetic fields from the inductors/transformers within, but detectable amounts do not appear to be emanating from DC and AC wires themselves - and that's the important part as they would otherwise be acting as antennas.

Perhaps the most important part of this modification is the fact that any bypass capacitors are placed on the "quiet" (not the inverter) side of the filtering inductances and that these bypass capacitors are connected, with short leads, to a large, common-point ground - namely the case of the inverter.  If any of the "ground" leads had been more than an inch or two long, it's likely that the impedance of it would have reduced the efficacy of the filtering - but the case, being a solid chunk of extruded aluminum, forms a nice, low-impedance tie point - effectively a single-point ground, preventing an RF current differential between the DC input and AC output leads.

* * *

Links to other articles about power supply noise reduction found at ka7oei.blogspot.com:

• A high-current DC noise filter for UPS or RV use.  What if you have an inverter or UPS that is large - as in the multi-kilowatt class?  This article describes quieting one such device.