{"id":86,"date":"2015-09-01T13:07:11","date_gmt":"2015-09-01T16:07:11","guid":{"rendered":"http:\/\/va1der.ca\/?page_id=86"},"modified":"2017-09-20T23:51:19","modified_gmt":"2017-09-21T02:51:19","slug":"project-archie","status":"publish","type":"page","link":"https:\/\/va1der.ca\/index.php\/project-archie\/","title":{"rendered":"Project &#8220;Archie&#8221; &#8211; The SB-200 Restoration Project"},"content":{"rendered":"<h2><span id=\"Introduction\">Introduction<\/span><\/h2><div id=\"toc_container\" class=\"toc_wrap_right toc_transparent no_bullets\"><p class=\"toc_title\">Contents<\/p><ul class=\"toc_list\"><li><a href=\"#Introduction\"><span class=\"toc_number toc_depth_1\">1<\/span> Introduction<\/a><\/li><li><a href=\"#Mission_Statement\"><span class=\"toc_number toc_depth_1\">2<\/span> Mission Statement<\/a><\/li><li><a href=\"#Refurbishment_Philosophy\"><span class=\"toc_number toc_depth_1\">3<\/span> Refurbishment Philosophy<\/a><\/li><li><a href=\"#Methodology\"><span class=\"toc_number toc_depth_1\">4<\/span> Methodology<\/a><\/li><li><a href=\"#Subsystems\"><span class=\"toc_number toc_depth_1\">5<\/span> Subsystems<\/a><ul><li><a href=\"#Power_Supply\"><span class=\"toc_number toc_depth_2\">5.1<\/span> Power Supply<\/a><ul><li><a href=\"#Circuit_Description\"><span class=\"toc_number toc_depth_3\">5.1.1<\/span> Circuit Description<\/a><ul><li><a href=\"#Power_Input\"><span class=\"toc_number toc_depth_4\">5.1.1.1<\/span> Power Input:<\/a><\/li><li><a href=\"#Power_Output\"><span class=\"toc_number toc_depth_4\">5.1.1.2<\/span> Power Output:<\/a><\/li><li><a href=\"#Balance_Resistors\"><span class=\"toc_number toc_depth_4\">5.1.1.3<\/span> Balance Resistors<\/a><\/li><\/ul><\/li><li><a href=\"#Common_Power_Mods\"><span class=\"toc_number toc_depth_3\">5.1.2<\/span> Common Power Mods<\/a><ul><li><a href=\"#Harbach_PM-200_Power_Board\"><span class=\"toc_number toc_depth_4\">5.1.2.1<\/span> Harbach PM-200 Power Board<\/a><\/li><li><a href=\"#Harbach_SS-201_Soft_Start\"><span class=\"toc_number toc_depth_4\">5.1.2.2<\/span> Harbach SS-201 Soft Start<\/a><\/li><\/ul><\/li><li><a href=\"#Refurbishment_Plan\"><span class=\"toc_number toc_depth_3\">5.1.3<\/span> Refurbishment Plan<\/a><ul><li><a href=\"#HV_Board\"><span class=\"toc_number toc_depth_4\">5.1.3.1<\/span> HV Board<\/a><\/li><li><a href=\"#Balance_Resistor_Redux\"><span class=\"toc_number toc_depth_4\">5.1.3.2<\/span> Balance Resistor Redux<\/a><\/li><\/ul><\/li><\/ul><\/li><\/ul><\/li><li><a href=\"#The_rest_is_UNDER_CONSTRUCTION\"><span class=\"toc_number toc_depth_1\">6<\/span> The rest is UNDER CONSTRUCTION<\/a><\/li><\/ul><\/div>\n\n<p>Welcome to Project &#8220;Archie&#8221;.\u00a0 This is the project to refurbish my Heathkit SB-200 HF linear amplifier.\u00a0 As the purpose of an HF linear amplifier is to make your signal just as loud as practicable, I thought it fitting to name it after the loudest-mouthed S.O.B. I could think of&#8230; <span title=\"As a kid I hated All in the Family exactly because of how bigoted, obnoxious, and annoying the Archie character was, but when I think of loud, that's the first thing that comes to mind\">Archie Bunker<\/span>.<\/p>\n<p>This is my first major hands-on ham project.\u00a0 A linear amplifier is one of those things where the circuitry is fairly simple, but where there has been a whole lot of thought that has had to go into the design before hand.\u00a0 As such, this project is as much about learning HF and dusting the cobwebs off general electronics knowledge as it is about actually refurbishing the hardware.\u00a0 The SB-200 is perfect for this on both fronts.\u00a0 Heathkits were intended to learn from.\u00a0 It was originally hand-assembled, and has instructions on literally how to piece it together from spare parts.\u00a0 This will be fun.<\/p>\n<h2><span id=\"Mission_Statement\">Mission Statement<\/span><\/h2>\n<blockquote><p>Learn about HF linear amplifiers by refurbishing &#8220;Archie&#8221;, my Heathkit SB-200 linear amplifier; bring it to the best possible operational condition with such modifications as required to make it safe and interoperable with other equipment while maintaining the spirit of its design and retaining as much of the original construction as possible.<\/p><\/blockquote>\n<h2><span id=\"Refurbishment_Philosophy\">Refurbishment Philosophy<\/span><\/h2>\n<p>Before I go into what I intend to do, a little bit on what my philosophy is going to be for this and other vintage hardware.<\/p>\n<p>This little Heathkit gem is a legend.\u00a0 First introduced in 1964, it was produced continuously (with one adjustment and model number bump in 1978 to comply with FCC rules) all the way until 1984.\u00a0 Seriously, what model of any piece of gear still used and relevant today do you know of that was made for twenty years?\u00a0 Twenty straight years of production (and sales) for any single design is a testament to its engineering.<\/p>\n<p>The reason I point this out, is that there are a great many SB-200 projects out there that advocate significant alterations to the amplifier&#8217;s basic design.\u00a0 It&#8217;s almost a given that most people will replace the high-voltage power board with a Harbach one or add their soft-start module (more on this later).\u00a0 Many people complain about the parasitic suppressor, and still others have partially or even mostly gutted the whole thing to give it 160m or replace the tube finals with something like a GI-7B.\u00a0 First of all, I&#8217;m a fairly new ham, so I&#8217;ll freely admit that a some of these changes, like the parasitic suppressor, I only partially understand the theory of.\u00a0 But secondly, even though my understanding of some of these topics is imperfect, it&#8217;s already clear to me that many of the changes some people advocate\u00a0 are ones <em>they<\/em> (at best) only partially understand.\u00a0 Many of them are advocating stamp-pad changes to this amplifier based on what they have seen done to other amplifiers.\u00a0 I have come the opinion that the Heathkit engineers of the 1960&#8217;s, 1970&#8217;s and 1980&#8217;s knew more about most of the design aspects of the SB-200 than the vast majority of hams today.\u00a0 If a hairpin suppressor really would have worked better, I suspect they would have put out the upgrade.<\/p>\n<p>Also, while I am not an &#8220;original parts purist&#8221;, I think it&#8217;s a travesty to see someone buy a vintage piece of ham radio history and just gut it.\u00a0 It&#8217;s like buying a 1964 Corvette and cramming a 6.2l LT4 engine in it just to say they did it.\u00a0 Sure, there is a certain coolness factor in saying &#8220;I did it&#8221;, so ten out of ten for style, but (with apologies to Douglas Adams), negative several million out of ten for destroying a work of art to do it.\u00a0 There are already too many irreplaceable pieces being taken out of circulation, we don&#8217;t need more of them taken out by people who do that just to see if they can.<\/p>\n<p>So, my refurbishment policy (which is visible in the mission statement) is this:<\/p>\n<ul>\n<li>Replace whatever components require swap out due to age or wear<\/li>\n<li>Not to be too slavish to the original design, but to hold fast to its spirit while making whatever upgrades are required either to make it safer or to make it work with the rest of my equipment<\/li>\n<li>If while doing the above I can make performance upgrades due to the improved technology of modern components, do so.\u00a0 Otherwise, if it ain&#8217;t broke, don&#8217;t fix it.<\/li>\n<\/ul>\n<h2><span id=\"Methodology\">Methodology<\/span><\/h2>\n<p>My methodology is simple enough.\u00a0 Follow the schematic.\u00a0 Start at the beginning and go step-by-step through to the end and ensure everything works.\u00a0 The unit I bought was already a working unit, with everything original except the final tubes.\u00a0 No mods, no upgrades.\u00a0 It&#8217;s a little down on 15m, but other than that, it was working when I bought it.\u00a0 This project is really about learning how the various subsystems work together to make a linear amplifier and about dusting off my general knowledge of electronics.<\/p>\n<h2><span id=\"Subsystems\">Subsystems<\/span><\/h2>\n<p>Like most pieces of electronics, the SB-200 is made of several different subsystems.\u00a0 Some of them are clearly separate, some of them could be divided several different ways and so the way I&#8217;m looking at it is just my own organization scheme:<\/p>\n<ul>\n<li>Power Supply &#8211; the 2400v power supply<\/li>\n<li>Input &#8211; RF input and keying<\/li>\n<li>Output &#8211; From tubes to antenna out<\/li>\n<li>Metering &amp; Misc &#8211; I&#8217;m looking at the metering circuits separately and throwing in miscellaneous bits here<\/li>\n<\/ul>\n<h3><span id=\"Power_Supply\">Power Supply<\/span><\/h3>\n<p>This is a critical subsystem.\u00a0 I suppose you can argue that they all are, but in this case, a failure of this circuit can cause catastrophic problems down the line.\u00a0 More especially because we&#8217;re working with high voltages.\u00a0 A failure here has the potential to be rather spectacular and also lethal.\u00a0 When I say 2400v is nothing to sneeze at, I&#8217;m actually serious &#8211; don&#8217;t sneeze if you&#8217;ve got a hand on a probe going in there.<\/p>\n<p>First of all I&#8217;m going to give a detailed description of the circuit, as much for my edification as yours.\u00a0 Then I&#8217;ll go into what changes are commonly done, why the two Harbach boards used here are a really bad idea, and what my intentions are power-wise for this refurb.<\/p>\n<h4><span id=\"Circuit_Description\">Circuit Description<\/span><\/h4>\n<h5><span id=\"Power_Input\">Power Input:<\/span><\/h5>\n<div id=\"attachment_146\" style=\"width: 310px\" class=\"wp-caption alignleft\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-Input-Stage.png\"><img aria-describedby=\"caption-attachment-146\" loading=\"lazy\" class=\"wp-image-146 size-medium\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-Input-Stage-300x215.png\" alt=\"SB-200 power supply input stage\" width=\"300\" height=\"215\" srcset=\"https:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-Input-Stage-300x215.png 300w, https:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-Input-Stage.png 986w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><\/a><p id=\"caption-attachment-146\" class=\"wp-caption-text\">SB-200 power supply input stage<\/p><\/div>\n<p>The original design works by taking line voltage and running it through transformer T1 to bump it up to ~800V RMS, or about ~1150V PtP.\u00a0 Nowadays with modern household voltages up around 120v in North America, this is closer to 1230V.\u00a0 From here on out I&#8217;m calling it 1200v.\u00a0 There are two equal primary coils on the transformer, able to be wired in series or parallel for either 110v or 220v operation.\u00a0 Each input leg of the transformer is protected by its own circuit breaker.\u00a0 This is necessary because the switching for series or parallel operation of the input primaries is done before the circuit breaker, which means if they are operated in parallel for 110v input then if the circuit breaker went off on one the other would still be energized without its own breaker.\u00a0 It could have been wired with the circuit breaker before the split and they could have gotten away with one, but I like this design the way it is.\u00a0 It works well for people in North America who choose to operate it on 220V.\u00a0 In North America, each leg (black and red) of a 220v line is hot with respect to neutral (white) and ground (green).\u00a0 This means a single pole circuit breaker on a 220v line can trip and you still have live power through the other leg with respect to ground.\u00a0 The two-breaker design ensures that whether you wire it for 110v or 220v either in North America or internationally, you are equally well protected.<\/p>\n<h5><span id=\"Power_Output\">Power Output:<\/span><\/h5>\n<p>T1 has three secondaries, one 800v for the main high voltage power, which we are interested in here, and two others, 6.3v and 110v for the heater\/lamps and tube bias which I&#8217;ll cover later.\u00a0 The 800v output goes to a modified <a href=\"https:\/\/en.wikipedia.org\/wiki\/Voltage_doubler#Delon_circuit\">Delon circuit<\/a> for voltage doubling.\u00a0 This is what we see on the high voltage board, and this is the board that is commonly replaced with a <a href=\"http:\/\/harbachelectronics.com\/shop\/heathkit-sb-200-sb-201\/pm-200-replacement-power-supply-module\/\" target=\"_blank\" rel=\"noopener noreferrer\">Harbach one<\/a>.<\/p>\n<div id=\"attachment_142\" style=\"width: 257px\" class=\"wp-caption alignleft\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-HV-e1441129944268.png\"><img aria-describedby=\"caption-attachment-142\" loading=\"lazy\" class=\"wp-image-142 size-medium\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-Schematic-Power-Supply-HV-247x300.png\" alt=\"SB-200 power supply high-voltage circuit\" width=\"247\" height=\"300\" \/><\/a><p id=\"caption-attachment-142\" class=\"wp-caption-text\">SB-200 power supply high-voltage circuit<\/p><\/div>\n<p>The voltage doubler works by charging two banks of three capacitors each to the maximum peak-to-peak voltage. When the transformer output line labelled &#8220;RED&#8221; is positive, it charges C4 &#8211; C6.\u00a0 When it&#8217;s negative, and RED-YEL is positive, then it charges C7 &#8211; C9.\u00a0 Eight diodes and three capacitors are used in series in each bank so as to handle the full peak-to-peak voltage voltage.\u00a0 A resistor (R5 &#8211; R10) is wired in parallel with each capacitor for two reasons, to act as a bleeder when power is off and as a balancing resistor for series charging.\u00a0 More on the balancing resistors below.\u00a0 Rounding out the circuit are resistors R11, R12, and R13.\u00a0 R11 forms a voltage divider with R10 to tap 10V off of the last capacitor&#8217;s 400V.\u00a0 This voltage is used for the Automatic Level Control feedback from the amplifier back to the exciter.\u00a0 R12 is similar &#8211; it forms a voltage divider between the entire capacitor network and ground and is used to measure the plate current without measuring bleeder current.\u00a0 More on this in metering.<\/p>\n<h5><span id=\"Balance_Resistors\">Balance Resistors<\/span><\/h5>\n<p>As mentioned, the capacitors are charged in two series banks of three capacitors each to handle the 1200v per bank.\u00a0 When you are charging capacitors in series, you need to account for the leakage current through them.\u00a0 Every capacitor leaks a little current, and electrolytic capacitors more than others.\u00a0 If the leakage currents were identical, then you wouldn&#8217;t need to worry.\u00a0 But that&#8217;s not the case.\u00a0 Each capacitor (even ones produced in the same batch) can have a different leakage current, which means when they are wired in series and you are forcing the same current through all three, they will all rise to different voltages.\u00a0 Let&#8217;s imagine that two have high leakage currents and one has a low leakage.\u00a0 Higher current means less &#8220;effective&#8221; resistance, and lower resistance means a lower voltage drop.\u00a0 But that 1200v total has to go somewhere, so if two capacitors are taking proportionally lower, then the one with the low leakage current will be taking proportionally more and may exceed its maximum voltage.\u00a0 So the balancing resistors essentially put an intentional leakage past each of them that is relatively high compared to the actual leakage. This balances out the voltages.<\/p>\n<div id=\"attachment_150\" style=\"width: 310px\" class=\"wp-caption alignright\"><img aria-describedby=\"caption-attachment-150\" loading=\"lazy\" class=\"size-medium wp-image-150\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-HV-Board-Original-300x212.jpg\" alt=\"SB-200 - Original high-voltage board\" width=\"300\" height=\"212\" srcset=\"https:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-HV-Board-Original-300x212.jpg 300w, https:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/SB-200-HV-Board-Original.jpg 814w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><p id=\"caption-attachment-150\" class=\"wp-caption-text\">SB-200 &#8211; Original high-voltage board<\/p><\/div>\n<p>Back in 1964 capacitors had larger leakage currents which required even more current through their balancing resistors.\u00a0 R5 &#8211; R10 are each 30K\u2126 resistors, and they are rated at 7w because at 400V per capacitor, a 30K\u2126 balancing resistor burns through 5\u2153W.\u00a0 Now I have no problem with Heathkit&#8217;s logic here, it was necessary after all.\u00a0 What I have a problem with is where they put them.\u00a0 Take a look at the photo of my board. At 5\u2153W per resistor, that&#8217;s 32W of heat being produced right beside those capacitors. My capacitors are, luckily, all in really good shape &#8211; I don&#8217;t think my unit was used much,What I couldn&#8217;t really show, due to the angle, was that each capacitor is about 7cm (2\u00be&#8221;) tall.\u00a0 And they are black. Which all means the resistors are radiating heat on the bottom of the board and that the capacitors are just soaking up because there is no ventilation on this side of the unit.<\/p>\n<div id=\"attachment_152\" style=\"width: 160px\" class=\"wp-caption alignright\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/WA3DSP-SB-200-HV-Board-Original.jpg\"><img aria-describedby=\"caption-attachment-152\" loading=\"lazy\" class=\"size-thumbnail wp-image-152\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/WA3DSP-SB-200-HV-Board-Original-150x150.jpg\" alt=\"Leaked electrolyte in a bad SB-200 board, courtesy of Doug Crompton (WA3DSB)\" width=\"150\" height=\"150\" \/><\/a><p id=\"caption-attachment-152\" class=\"wp-caption-text\">Leaked electrolyte in a bad SB-200 board, courtesy of Doug Crompton (WA3DSB)<\/p><\/div>\n<p>These little space heaters are the reason for most of the capacitor failures in these units.\u00a0 One of two things should have been done here.\u00a0 A better design would have been to mount the board higher, hang the capacitors off the bottom and having the resistors on top.\u00a0 This would keep the board between the resistors and capacitors and have the heat free to rise up away.\u00a0 Even a little heat-shield divider placed between the resistors and capacitors to protect them from the heat would have been better than nothing.\u00a0 Something to separate that heat from the capacitors.\u00a0 The capacitors on my unit seem to be in good condition &#8211; excellent actually, but you can see here on the board of a unit that is a little older than mine what effect those lovely heaters have had.<\/p>\n<h4><span id=\"Common_Power_Mods\">Common Power Mods<\/span><\/h4>\n<p>There are two very common mods people purchase for the SB-200&#8217;s power supply system, both of them from Harbach.<\/p>\n<h5><span id=\"Harbach_PM-200_Power_Board\">Harbach PM-200 Power Board<\/span><\/h5>\n<div id=\"attachment_157\" style=\"width: 160px\" class=\"wp-caption alignright\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/WA3DSP-SB-200-Harbach-Power-Board-V2-.jpg\"><img aria-describedby=\"caption-attachment-157\" loading=\"lazy\" class=\"wp-image-157 size-thumbnail\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/WA3DSP-SB-200-Harbach-Power-Board-V2--150x150.jpg\" alt=\"WA3DSP SB-200 - Harbach Power Board V2+\" width=\"150\" height=\"150\" \/><\/a><p id=\"caption-attachment-157\" class=\"wp-caption-text\">Assembled Harbach PM-200 board, courtesy of Doug Crompton (WA3DSB)<\/p><\/div>\n<p>One of these is their PM-200 replacement high-voltage board, shown at right.\u00a0 As I understood it, one of the original purposes of this board was to give you something with modern capacitors having much lower leakage currents so that the balance resistors could be increased in value and reduce the quiescent power draw of the unit.\u00a0 That&#8217;s what most SB-200 project pages state.\u00a0 And on <a href=\"http:\/\/harbachelectronics.com\/shop\/heathkit-sb-200-sb-201\/pm-200-replacement-power-supply-module\/\" target=\"_blank\" rel=\"noopener noreferrer\">Harbach&#8217;s page<\/a> for the module they tell you they are using &#8220;new 82K\u03a9 3W bleeder resistors&#8221; &#8211; but what they don&#8217;t tell you is that they are using two of them in parallel for each capacitor.\u00a0 That&#8217;s 41k\u2126, burning 3.9 watts per capacitor for 23.4 watts total &#8211; and again, all of the resistors are placed right beside one bank of the capacitors.\u00a0 So the Harbach board is no real benefit at all.\u00a0 They are either using cheap capacitors with very high leakage currents, or more likely are basing their balance resistor values on outdated thinking (see below).\u00a0 Harbach&#8217;s capacitors are physically smaller than the originals, and the resistors are burning <em>slightly<\/em> less power, both of which mean the capacitors won&#8217;t get quite as hot as on the original board. But unless your original board is physically cracked, I see nothing about the Harbach PM-200 board that is tempting.<\/p>\n<div id=\"attachment_162\" style=\"width: 160px\" class=\"wp-caption alignright\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/K4LAJ-SB-200-New-Power-Board.jpg\"><img aria-describedby=\"caption-attachment-162\" loading=\"lazy\" class=\"size-thumbnail wp-image-162\" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/K4LAJ-SB-200-New-Power-Board-150x150.jpg\" alt=\"K4LAJ's SB-200 HV power board - note the balance resistor values and placement\" width=\"150\" height=\"150\" \/><\/a><p id=\"caption-attachment-162\" class=\"wp-caption-text\">K4LAJ&#8217;s SB-200 HV power board &#8211; note the balance resistor values and placement<\/p><\/div>\n<p>If you are in the market for a brand new board, then a better solution might be the one created by fellow ham Don Van Wagner (K4LAJ).\u00a0 He operates a small-volume business under the eBay seller name vanwagnd112.\u00a0 He sells fully populated and assembled power boards.\u00a0 Note that he&#8217;s using 100K\u2126 balancing resistors (burning 1.6 watts each for a total of 9.6 watts) and that he&#8217;s designed it so each resistor is located beside the capacitor it balances.\u00a0 There is less waste heat, and that heat is more evenly distributed.\u00a0 I will note, though, that while I can say that the 100K\u2126 is more in line with the leakage current in modern capacitors (and even less than I&#8217;ll be using &#8211; see below), without seeing the data sheet for these capacitors I can&#8217;t say for that this is an appropriate value in this case.\u00a0 It&#8217;s certainly a much better layout.<\/p>\n<h5><span id=\"Harbach_SS-201_Soft_Start\">Harbach SS-201 Soft Start<\/span><\/h5>\n<p>The Harbach SS-201 Soft Start board is the other common (almost ubiquitous) mod that SB-200 owners are installing.\u00a0 My feelings on this board have changed over the last year.\u00a0 In draft versions of this article I wrote that the board was unsafe.\u00a0 And while I still don&#8217;t think the circuit is the best design, I&#8217;m not prepared any more to actively call it unsafe.\u00a0 I will say I believe it unnecessary.<\/p>\n<p>The Harbach soft start board works by running two 10\u2126 10W (for 110V operation) resistors in series with the incoming AC mains power for what Harbach says is &#8220;a fre milliseconds&#8221; when the device is powered up, then switches them out and runs AC mains power directly in.\u00a0 The intention is to limit the inrush of power in the amp (ie: filaments) and power it up gracefully.\u00a0 Two relays do the switching.\u00a0 When the relays are in their normally closed configuration, the resistors are switched in.\u00a0 When the relays are powered, then the resistors are switched out.\u00a0 That makes me uncomfortable because the most common failure mode for any relay is for it to fail closed.\u00a0 Which means if one of those relays fails, then the resistor will be left in series with mains power switched in. If a relay does fail, what happens?<\/p>\n<ol>\n<li>One or both relays fail.<\/li>\n<li>110V power is now directed through the soft-start 10W resistors<\/li>\n<li>The amp is not being keyed, so the power supply is drawing between 10-30 watts depending on the power supply board installed.\u00a0 Worst case, 30 watts.\u00a0 A 10 watt wire wound resistor won&#8217;t let go with 30 watts going through it (at least not for a long time), but it will get very hot.\u00a0 What effect that has depends on where in the case the board is mounted.\u00a0 It likely won&#8217;t get hot enough to start a fire at 30 watts, but it could easily melt through wiring.\u00a0 This could be catastrophic right there, but lets say not.\u00a0 Other than potentially melting wiring, unless your front meter is switched into the HV line, there won&#8217;t be any real indication that the relay has failed.\u00a0 The lights will come on.\u00a0 The fan will come on.\u00a0 But then at some point&#8230;<\/li>\n<li>The operator will key the exciter&#8217;s mike.\u00a0 Or, more likely, key the CW to tune up the amp.\u00a0 At this point, I work out that the there will be about 380 watts of power dissipating through those 10 watt resistors.\u00a0 The outcome is a foregone conclusion.<\/li>\n<\/ol>\n<p>I&#8217;ve only seen 10W resistors let go a couple times, and both times it was like a small pyrotechnic.\u00a0 They are not designed to be fuses, letting go gracefully when too much power flows through them.\u00a0 They are designed to be tough and try to dissipate just as much heat as they possibly can.\u00a0 They have a strong ceramic housing, which in this case just acts as a pressure vessel.\u00a0 By the end of the scenario, they will already have been very hot, and possibly have already melted wiring and plastic housing on the nearby relays, so when they let go they can spread hot melted plastic and ceramic and wire bits all over.\u00a0 And, by design, this board is mounted very close to the HV power board and close by the input terminals where AC mains power is coming into the machine.<\/p>\n<p>How likely is it that one of those relays will fail?\u00a0 Admittedly not bloody likely, which is why I&#8217;ve backed off saying these boards are unsafe.\u00a0 Most relays are rated for millions of cycles.\u00a0 I just find it ironic that most people who own these amps will install a couple of protection diodes across the meter just in the off chance that something does fail (because the meter is irreplaceable), but they then seem to be blissfully happy to install a soft start board where they are <em>adding<\/em> a point of failure there is no protection for.\u00a0 The way I see it, anyone with this board in their amp is one Chinese-manufactured relay failure away from a fire in their amplifier.<\/p>\n<p>That is an unlikely scenario.\u00a0 So let&#8217;s analyze what goes on during a typical &#8220;soft&#8221; start. What does a soft start board buy you anyway?\u00a0 Its intended purpose is to reduce inrush current &#8220;for a few milliseconds&#8221; at startup, thus reducing strain on your tube filaments and on the HV board&#8217;s rectifier diodes.\u00a0 I have read about a lot of SB-200&#8217;s with failed components, I have yet to see a report for one where the rectifier diodes let go.\u00a0 They have a lot of surge capacity.\u00a0 And for the tubes, filament failure is by far one of the least common failure modes for a 572b.\u00a0 Now why is that?\u00a0 After all, studies have shown that soft start modules in general <em>do<\/em> help prevent vacuum tube filament failures, so why do the 572b filaments (and power amplifier tube filaments in general) so seldomly fail?\u00a0 Filament inrush current, while theoretically very large, is in practice limited by the transformer.\u00a0 In an SB-200 (and almost every other similar amp) the filament inrush is competing against the large voltage doubler power capacitors for that initial surge current.\u00a0 A cold 572b filament measures at about 0.3\u2126, which means it wants to draw 21A surge current at startup.\u00a0 That drops to less than twice the nominal current of 4A after ten milliseconds and the current inrush period is essentially ended.\u00a0 It is guaranteed that for that for at last the first 16.7 milliseconds (the time of one power cycle) your HV power board capacitors will have far less effective resistance than the tube filaments and will essentially be drawing all the current the transformer can produce.\u00a0 Which means for far longer than the &#8220;few milliseconds&#8221; that a Harbach soft start board is active at startup, your HV board power capacitors in concert with your transformer are acting to limit how much current your tube filaments are getting anyway.\u00a0 In fact, the filament surge period isn&#8217;t even over by the time a Harbach board switches its current limiting resistors out of the circuit.<\/p>\n<p>So my conclusion for any tube amplifier with large capacitors in the power board, the Harbach soft start board is completely unnecessary, and essentially ineffective.\u00a0 Mommy medicine that will make you feel better for putting it in, but does nothing for your amp.<\/p>\n<h4><span id=\"Refurbishment_Plan\">Refurbishment Plan<\/span><\/h4>\n<h5><span id=\"HV_Board\">HV Board<\/span><\/h5>\n<p>My HV power board is in excellent shape.\u00a0 When operating, there is between 400-410 volts measured across each capacitor.\u00a0 That is a very small spread and it tells me the capacitors are in good shape relative to each other.\u00a0 I <strong>hate<\/strong> the idea of pulling out and replacing perfectly good, and original, components from a working piece of gear.\u00a0 However, as noted earlier, this is a critical board and I&#8217;m just not comfortable with leaving resistor-cooked capacitors that are no less than 30 years old sitting in it.\u00a0 This amp isn&#8217;t going to sit idle &#8211; it&#8217;s going to be used, and well used.\u00a0 I&#8217;m not going to jeopardize everything else in the unit.<\/p>\n<p>Before I found Don Van Wagner&#8217;s excellently designed board, my plan had been to simply replace the capacitors and required resistors <em>in situ<\/em> on the original board.\u00a0 After finding it, though, I came up with a new plan.\u00a0 Having a fully stock SB-200 in working order is becoming rarer, so rather than pull good components off the old board as a just in case, I&#8217;m going to replace the board outright and keep the old one fully intact.\u00a0 That gives me the option of returning the amp to its stock configuration some time in the future.<\/p>\n<div id=\"attachment_168\" style=\"width: 251px\" class=\"wp-caption alignright\"><a href=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/Snap-In-HV-Board-Fit.svg\"><img aria-describedby=\"caption-attachment-168\" loading=\"lazy\" class=\"wp-image-168 \" src=\"http:\/\/va1der.ca\/wp-content\/uploads\/2015\/09\/Snap-In-HV-Board-Fit.svg\" alt=\"\" width=\"241\" height=\"108\" \/><\/a><p id=\"caption-attachment-168\" class=\"wp-caption-text\">New capacitor placement on old board, shown from underneath<\/p><\/div>\n<p>The old plan would have worked fine, so if you have a board that already has bad component, then by all means refurbish it.\u00a0 I don&#8217;t think I would have had to do anything drastic to the original board to make new snap-in capacitors work, as shown in the diagram.\u00a0 I would have needed only jumper the negative lead.\u00a0 You will still end up with the resistors all on one side, but there is a great solution for the heat issue.\u00a0 Technology to the rescue.<\/p>\n<h5><span id=\"Balance_Resistor_Redux\">Balance Resistor Redux<\/span><\/h5>\n<p>This section ended up being very long.\u00a0 When doing research on balance resistors I encountered\u00a0 a lot of rules of thumb and old thinking and had to dig.\u00a0 I spawned off the explanation into a paper on <a href=\"http:\/\/va1der.ca\/index.php\/balance-resistors-for-series-capacitors\/\">balance resistors for series capacitors<\/a>.\u00a0 Condensing this down, the formula I recommend for balance resistors is:<\/p>\n<blockquote><span class=\"wp-katex-eq\" data-display=\"false\">R_{balance} = \\frac{NV_{rate}-V_{bus}}{I_{\\Delta bleed}}<\/span><\/blockquote>\n<p>Where:<\/p>\n<ul>\n<li><span class=\"wp-katex-eq\" data-display=\"false\">N<\/span> is the number of capacitors in series<\/li>\n<li><span class=\"wp-katex-eq\" data-display=\"false\">V_{rated}<\/span> is the rated maximum voltage for any one capacitor<\/li>\n<li><span class=\"wp-katex-eq\" data-display=\"false\">V_{bus}<\/span> is the bus voltage &#8211; the expected voltage across the whole series of capacitors<\/li>\n<li><span class=\"wp-katex-eq\" data-display=\"false\">I_{\\Delta bleed}<\/span> Bleed current difference in \u00b5A<\/li>\n<\/ul>\n<p>You end up with <span class=\"wp-katex-eq\" data-display=\"false\">R_{balance}<\/span> in megohms<\/p>\n<p>This formula calculates the total voltage &#8220;headroom&#8221; you have in the series bank of capacitors series and divides it by an estimation of the <em>difference<\/em> in bleed current.\u00a0 I used Cornell Dubilier&#8217;s formula for estimating this:<\/p>\n<blockquote><span class=\"wp-katex-eq\" data-display=\"false\">I_{\\Delta bleed} = 0.0015CV_{bus}<\/span><\/blockquote>\n<p>I ordered six 390<span class=\"wp-katex-eq\" data-display=\"false\">\/mu<\/span>F 500V Nippon Chemicon &#8220;LXS&#8221; series capacitors from Ali Express.\u00a0 I got the 500V rated capacitors since they are common enough now so as not to be expensive, and they give a lot of voltage headroom.\u00a0 A voltage doubler works by charging two banks of capacitors independently and then adding them together for discharge.\u00a0 They are being charged in two banks of three, so this works in the above formulae as:<\/p>\n<blockquote><span class=\"wp-katex-eq\" data-display=\"false\">R_{balance} = \\frac{3 \\cdot 500V - 1250V}{.0015 \\cdot 390{\\mu}F \\cdot 1250V} = 0.341M\\Omega<\/span><\/blockquote>\n<p>Even adding in a safety factor of two (and then rounding to give a power burn of an even one watt), the 160K\u03a9 value I chose is less than what the old rules of thumb state.<\/p>\n<h2><span id=\"The_rest_is_UNDER_CONSTRUCTION\">The rest is UNDER CONSTRUCTION<\/span><\/h2>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>IntroductionContents1 Introduction2 Mission Statement3 Refurbishment Philosophy4 Methodology5 Subsystems5.1 Power Supply5.1.1 Circuit Description5.1.1.1 Power Input:5.1.1.2 Power Output:5.1.1.3 Balance Resistors5.1.2 Common Power Mods5.1.2.1 Harbach PM-200 Power Board5.1.2.2 Harbach SS-201 Soft Start5.1.3 Refurbishment Plan5.1.3.1 HV Board5.1.3.2 Balance Resistor Redux6 The rest is UNDER &hellip; <a href=\"https:\/\/va1der.ca\/index.php\/project-archie\/\">Continue reading <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"open","ping_status":"open","template":"","meta":[],"_links":{"self":[{"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/pages\/86"}],"collection":[{"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/comments?post=86"}],"version-history":[{"count":32,"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/pages\/86\/revisions"}],"predecessor-version":[{"id":604,"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/pages\/86\/revisions\/604"}],"wp:attachment":[{"href":"https:\/\/va1der.ca\/index.php\/wp-json\/wp\/v2\/media?parent=86"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}