{"id":441,"date":"2020-03-04T23:25:03","date_gmt":"2020-03-04T20:25:03","guid":{"rendered":"https:\/\/tomorrow82.ru\/?page_id=441"},"modified":"2020-03-04T23:25:03","modified_gmt":"2020-03-04T20:25:03","slug":"parts-list-for-one-channel","status":"publish","type":"page","link":"https:\/\/tomorrow82.ru\/?page_id=441","title":{"rendered":"Parts List for One Channel"},"content":{"rendered":"\n<p><strong>Version 4.4\/4.5 of the Amplifier<\/strong><\/p>\n\n\n\n<p><strong>Transistors<\/strong> (Starred types are preferred.)<\/p>\n\n\n\n<p>I tend to prefer Motorola if available. Motorola has spun off its \nsemiconductor division, and it is now called On Semiconductor. In the \nprocess, they have discontinued many products. I have seen some mystery \nbrand transistors that have really bad characteristics. If the \ntransistor is labeled with a name brand manufacturer, it should give no \nproblems. For Q1 &#8212; Q11, the first NPN type should be used if the first \nPNP type is used, etc. This also applies to Q16 &#8212; Q17 and to Q18 &#8212; Q21. I\n have seen students use all of the types of transistors that I have \nlisted below with no problems. If you look at the specs, however, you \nwill see why I prefer the starred types. Transistors can be substituted,\n provided the substitutes have similar specs to the ones below.<\/p>\n\n\n\n<p>I have been told that the MPS8099\/MPS8599 are no longer a \n&#171;complementary&#187; pair because the PNP is discontinued.  ON Semiconductor \nstates that the direct replacement for the MPS8599 is a MPSA56. In March\n 2003, I was told that the MPS8599 is back. ON Semiconductor is now \nsupplying them.<\/p>\n\n\n\n<p>Other possibilities for the MPS8099\/MPS8599 are the ZTX651\/ZTX751 which are available from Digi-Key.<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q1, Q2, Q5, Q7, Q9, Q10<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> at 25<sup>o<\/sup> C ambient<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MPS8099*<\/td><td>80 V<\/td><td>0.2 A<\/td><td>350 mW<\/td><td>100<\/td><td>150 MHz<\/td><\/tr><tr><td>MPS8098<\/td><td>60 V<\/td><td>0.2 A<\/td><td>350 mW<\/td><td>100<\/td><td>150 MHz<\/td><\/tr><tr><td>MPSA06<\/td><td>80 V<\/td><td>0.5 A<\/td><td>300 mW<\/td><td>50<\/td><td>50 MHz<\/td><\/tr><tr><td>MPSA05<\/td><td>60 V<\/td><td>0.5 A<\/td><td>300 mW<\/td><td>50<\/td><td>50 MHz<\/td><\/tr><tr><td>2N5210<\/td><td>50 V<\/td><td>0.05 A<\/td><td>310 mW<\/td><td>200<\/td><td>30 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q3, Q4, Q6, Q8, Q11<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> at 25<sup>o<\/sup> C ambient<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MPS8599*<\/td><td>80 V<\/td><td>0.2 A<\/td><td>350 mW<\/td><td>100<\/td><td>150 MHz<\/td><\/tr><tr><td>MPS8598<\/td><td>60 V<\/td><td>0.2 A<\/td><td>350 mW<\/td><td>100<\/td><td>150 MHz<\/td><\/tr><tr><td>MPSA56<\/td><td>80 V<\/td><td>0.5 A<\/td><td>300 mW<\/td><td>50<\/td><td>50 MHz<\/td><\/tr><tr><td>MPSA55<\/td><td>60 V<\/td><td>0.5 A<\/td><td>300 mW<\/td><td>50<\/td><td>50 MHz<\/td><\/tr><tr><td>2N5087<\/td><td>50 V<\/td><td>0.05 A<\/td><td>310 mW<\/td><td>200<\/td><td>30 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q13, Q14<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> at 25<sup>o<\/sup> C ambient<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>2N3439*<\/td><td>350 V<\/td><td>1 A<\/td><td>1 W<\/td><td>40<\/td><td>15 MHz<\/td><\/tr><tr><td>2N3440<\/td><td>250 V<\/td><td>1 A<\/td><td>1 W<\/td><td>40<\/td><td>15 MHz<\/td><\/tr><tr><td>NTE396<\/td><td>350 V<\/td><td>1 A<\/td><td>1 W<\/td><td>30<\/td><td>15 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q12, Q15<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> at 25<sup>o<\/sup> C ambient<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>2N5415<\/td><td>200 V<\/td><td>1 A<\/td><td>1 W<\/td><td>30<\/td><td>15 MHz<\/td><\/tr><tr><td>2N5416*<\/td><td>300 V<\/td><td>1 A<\/td><td>1 W<\/td><td>30<\/td><td>15 MHz<\/td><\/tr><tr><td>NTE397<\/td><td>350 V<\/td><td>1 A<\/td><td>10 W<\/td><td>30<\/td><td>?<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p>I have not tried the NTE396 and NTE397 transistors. They are \nspecified to be a complementary pair. The spec sheet for the NTE396 has \ntwo power ratings &#8212; 1 W and 5 W. It is not clear why they specify two. \nThe spec sheet for the NTE397 only has one power rating and it does not \ngive the gain bandwidth product f<sub>T<\/sub>.<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q16<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> with case at 25<sup>o<\/sup> C<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MJE15030*<\/td><td>150 V<\/td><td>8 A<\/td><td>50 W<\/td><td>20<\/td><td>30 MHz<\/td><\/tr><tr><td>2N6474<\/td><td>120 V<\/td><td>4 A<\/td><td>40 W<\/td><td>15<\/td><td>4 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q17<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> with case at 25<sup>o<\/sup> C<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MJE15031*<\/td><td>150 V<\/td><td>8 A<\/td><td>50 W<\/td><td>20<\/td><td>30 MHz<\/td><\/tr><tr><td>2N6476<\/td><td>120 V<\/td><td>4 A<\/td><td>40 W<\/td><td>15<\/td><td>4 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q18, Q20<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> with case at 25<sup>o<\/sup> C<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MJ15003*<\/td><td>140 V<\/td><td>20 A<\/td><td>250 W<\/td><td>25<\/td><td>2 MHz<\/td><\/tr><tr><td>MJ15001<\/td><td>140 V<\/td><td>15 A<\/td><td>200 W<\/td><td>25<\/td><td>2 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<figure class=\"wp-block-table is-style-regular\"><table class=\"\"><tbody><tr><td>Q19, Q21<\/td><td>BV<sub>CEO<\/sub><\/td><td>I<sub>C(max)<\/sub><\/td><td>P<sub>D(max)<\/sub> with case at 25<sup>o<\/sup> C<\/td><td>h<sub>FE(min)<\/sub><\/td><td>f<sub>T(min)<\/sub><\/td><\/tr><tr><td>MJ15004*<\/td><td>140 V<\/td><td>20 A<\/td><td>250 W<\/td><td>25<\/td><td>2 MHz<\/td><\/tr><tr><td>MJ15002<\/td><td>140 V<\/td><td>15 A<\/td><td>200 W<\/td><td>25<\/td><td>2 MHz<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p>Do not use plastic insulating caps over Q18 through Q21. These will \nkeep the screws that hold the power transistors to their sockets from \nmaking contact to the transistor cases. The case is the collector and \nthere will be no power supply voltage on the collectors. I had a student\n use these caps, and R41 through R44 caught on fire on his board, and he\n blew Q16 and Q17.<\/p>\n\n\n\n<p>Apparently the 2N3439\/2N3440 and 2N5415\/2N5416 transistors are \ngetting difficult to find. A possible manufacturer who has them listed \non its web page is <a href=\"http:\/\/www.st.com\/stonline\/index.htm\">STMicroelectronics<\/a>. The STMicroelectronics transistors are available from <a href=\"http:\/\/www.arrow.com\">Arrow Electronics<\/a> and <a href=\"http:\/\/www.mouser.com\">Mouser Electronics<\/a>.\n I would be interested if anyone knows of other sources for these \ntransistors. I have been told that ON Semiconductor no longer makes \nthese and transistors with the ON Semiconductor label may be \ncounterfeit. The transistors are currently made by ST Microelectronics \nand Central Semiconductor.<\/p>\n\n\n\n<p>As of June 2000, <a href=\"http:\/\/www.radioshack.com\">RadioShack.com<\/a>  has all of the transistors at good prices. They can be ordered through  the web. You can see a list of their part numbers and prices.  In spring 2002, students told me that RadioShack.com was selling its  inventory at steep discounts and would soon be out of business. Formerly  Tech America, this is the parts division of Radio Shack and is not  connected with the retail stores.<\/p>\n\n\n\n<p>The 2N3439\/2N5415 pair is the one that I originally used for my \nprototype amplifiers. It is the complementary TO-5\/TO-39 case pair \nrecommended in the RCA Power Transistor Manual for audio amplifiers. For\n all practical purposes, the 2N3440 and 2N5416, respectively, are \nequivalent to the 2N3439 and the 2N5415 for the Low TIM amplifier.<\/p>\n\n\n\n<p>Q12 &#8212; Q15 can run a little warm. I have never had problems with these\n running too hot. With the Version 4.4\/4.5 amp, I have reduced the bias \ncurrents in these transistors just a little to keep them cooler. \nAlthough I don&#8217;t think they are needed, you can put TO-5 clip-on heat \nsinks on them. There is not much room on the circuit board for the heat \nsinks, so they must have small fins. Another type of heat sink which \nwill fit is in the shape of a flag (about 1 inch high by 3\/4 inch wide) \nwith rounded clips on one end that clip around the transistor. If you \ncan&#8217;t find them, they are easy to make from sheet metal flashing. The \nclips should make good mechanical contact to the transistor for good \nheat conduction.<\/p>\n\n\n\n<p><strong>Diodes<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>D1, D2, D3, D4, D11, D12 &#8212; 1N4004 (DigiKey 1N4004DICT-ND)<\/li><li>D5, D6, D7, D8, D9, D10 &#8212; 1N4148 (DigiKey 1N4148DICT-ND)<\/li><li>D13 through D16 &#8212; 1N5250B 20 V Zener (DigiKey 1N5250BDICT-ND, not used in Ver. 4.3)<\/li><\/ul>\n\n\n\n<p><strong>Capacitors<\/strong> (First type listed is preferred.)<\/p>\n\n\n\n<p>Some of the capacitor types specified below are ceramic. I have seen \nsome of these exhibit non-linear effects, so I don&#8217;t recommend them \nunless the mica types cannot be found. The ceramic dielectric in the \ncapacitors can exhibit a piezoelectric effect which causes the spacing \nbetween the plates to vary with applied voltage. This causes the \ncapacitance to vary with voltage. Indeed, on one occasion I could hear a\n ceramic capacitor &#171;sing&#187; when excited at its resonance frequency.<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>C1 &#8212; 390 pF mica (Arco DM15-391J or DigiKey 338-1057-ND, 0.25 inch hole spacing)<\/li><li>C2, C3, C15, C16, C23, C24, C25 &#8212; 0.1 uF, 100 V film (DigiKey P4725-ND, 0.4 inch or 0.2 inch hole spacing)<\/li><li>C7, C12, C17, C18 &#8212; 0.1 uF, 50 V film (DigiKey P4525-ND, 0.2 inch hole spacing)<\/li><li>C4, C5, C13, C14, C21, C22 &#8212; 100 uF, 63 V radial electrolytic (DigiKey P10343-ND, 0.2 inch hole spacing)<\/li><li>C6 &#8212; Either a single non-polar capacitor or two polar capacitors can\n be used for C6. I recommend the non-polar capacitor. In this case, use a\n 220 uF, 16 V bi-polar electrolytic in the holes for C6A \n(DigiKey\/Panasonic P1168-ND, 0.2 inch hole spacing) and solder a short \ncircuit jumper in the holes for C6B. (Failure to install this jumper \nwill result in the loss of all bass response.) For two polar capacitors,\n use a 330 uF, 16 V radial electrolytic for C6A and C6B (DigiKey \nP10246-ND).<\/li><li>C8 &#8212; 180 pF mica (Arco DM15-181J or DigiKey 338-1082-ND, 0.25 inch hole spacing)<\/li><li>C9 &#8212; 47 pF mica (Arco DM15-470J or DigiKey 338-1053-ND 5.9 mm hole \nspacing. The DigiKey 338-1084-ND has a 3 mm hole spacing but the leads \ncan be bent to fit the circuit board)<\/li><li>C10, C11 &#8212; 10 pF mica (Arco DM15-100J or DigiKey 338-1068-ND, 0.25 inch hole spacing)<\/li><li>C19, C20 &#8212; 0.01 uF, 50 V film (DigiKey P4513-ND, 0.2 inch hole spacing)<\/li><\/ul>\n\n\n\n<p><strong>Resistors<\/strong> (1\/4 W 5% carbon film or 1% metal film unless \nspecified otherwise. Please use an ohmmeter to check the value of all \nresistors before soldering them to the circuit board.)<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>R1 &#8212; 20 kohm<\/li><li>R2 &#8212; 2 kohm<\/li><li>R3 through R10 &#8212; 300 ohm<\/li><li>R11, R12, R27 &#8212; 1.2 kohm<\/li><li>R13, R14 &#8212; 2.2 kohm 1\/2 W (see note below, 3.6 kohm in Ver. 4.3)<\/li><li>R15, R16 &#8212; 12 kohm (10 kohm in Ver. 4.3)<\/li><li>R17 &#8212; R18 &#8212; 11 kohm<\/li><li>R19 &#8212; 1.1 kohm<\/li><li>R20 &#8212; 22 kohm<\/li><li>R21, R22 &#8212; 30 ohm<\/li><li>R23, R24 &#8212; 360 ohm<\/li><li>R25, R26 &#8212; 1 kohm<\/li><li>R28, R29 &#8212; 270 ohm<\/li><li>R30, R31 &#8212; 3.9 kohm 1\/2 W<\/li><li>R32, R33, R51 &#8212; 82 ohm<\/li><li>R34, R35 &#8212; 330 ohm (270 ohm in Ver. 4.3)<\/li><li>R36 &#8212; 220 ohm &#8212; If you have a problem with the V<sub>BE<\/sub>\n multiplier or the bias diodes, the bias voltage can go too high and \nthis resistor can smoke and possibly burn the circuit board. I recommend\n putting a 1\/8 inch long piece of insulation stripped from a piece of \nhookup wire on each lead of this resistor before soldering it to the \ncircuit board. The insulation will hold the resistor up off the circuit \nboard in case it burns. If you wish, you can use a 1\/2 watt resistor for\n R36.<\/li><li>R37 through R40 &#8212; 680 ohm (changed from 470 on 4\/4\/2)<\/li><li>R41 through R44 &#8212; 10 ohm 1\/2 W (changed from 3.3 ohms 7\/2\/00)<\/li><li>R45 through R48 &#8212; 0.33 ohm 5 W Wire Wound (DigiKey 0.33W-5-ND)<\/li><li>R49, R50 &#8212; 10 ohm, 2 W ceramic or carbon composition. Allen-Bradley \nmade the carbon composition resistors for many years, but they \napparently no longer make them. Ohmite is another manufacturer who is \nphasing out their carbon composition resistors in favor of ceramic \ncomposition. Digi-Key sells the Ohmite ceramic resistor. The part number\n is OY100K-ND. It measures 7\/8 inch long by 5\/16 inch in diameter. \nAnother resistor sold by Digi-Key is the ALSR5F-10-ND, 10 ohm, 5 W \nresistor.  It is about the same size as the OY100K-ND. Digi-Key also \nsells smaller size carbon film 2 W resistors. Although these can be \nused, I prefer the larger ones to wind the inductor L1 around. My \ncircuit boards are drilled to fit the ceramic and carbon film resistors.\n The holes must be enlarged for the older carbon composition units. We \nhave found that Ack Radio sells a Dale CW-5 10 ohm resistor (I believe \nit is rated at 5 watts) that can be used to wind the inductor on.<\/li><\/ul>\n\n\n\n<p>If you do not use the specified power supply voltages, you can \ncalculate the values for R13 and R14 from the formula R13 = R14 = (V &#8212; \n40)\/8.2, where V is the power supply voltage. For example, for V = 58 V,\n the formula gives R13 = R14 = 2.2 kohm. Use the nearest 5% resistor \nvalue. (For the Ver. 4.3 amplifier, calculate the values for R13 and R14\n from the formula R13 = R14 = (V &#8212; 38.2)\/5.42, where V is the power \nsupply voltage. For example, for V = 57.7 V, the formula gives R13 = R14\n = 3.6 kohm.)<\/p>\n\n\n\n<p><strong>Miscellaneous<\/strong><\/p>\n\n\n\n<ul class=\"wp-block-list\"><li>P1 &#8212; 2 kohm, 3\/8 inch square, single-turn, top-adjust, laydown version, cermet potentiometer (DigiKey 3386P-202-ND)<\/li><li>L1 &#8212; 10 to 12 turns #22 solid insulated wire wound tightly around  R49 and soldered to the leads of R49 where they emerge from the resistor  body. Click <a href=\"https:\/\/tomorrow82.ru\/wp-content\/uploads\/2020\/03\/l1r49.gif\">here<\/a>  to see an illustration. Solder one end of a piece of #22 solid wire to  one end of R49. Wind the wire tightly around R49 to form L1. Clamp the  windings to R49 with a small bench vise. Strip and solder the other end  of L1 to R49. Do not use stranded wire.<\/li><li>Use only stranded wire to wire the chassis. Do not use solid wire. I  recommend #18 or #20 stranded wire for all wiring except the wires that  run from the circuit boards to the bias diodes in the heat sinks and to  the bases of the output transistors. These are all low current leads  and I recommend #22 stranded wire for them. The holes in my circuit  boards are drilled for these size wires. Do not use wire that is too big  to fit into the holes in the circuit boards.<\/li><li>Power Switch &#8212; Digikey (EG1858-ND). This is a rocker switch which  has been used by a student and he told me that he had no problems with  it sticking. I have seen so many rocker switches purchased by students  stick that I hesitate to recommend one. Maybe this one is an exception.<\/li><li>Heat sinks for Q16 and Q17 are Aavid 5741B (DigiKey HS111-ND).<\/li><li>Heat sink for Q18 through Q21 is Wakefield  423K or equivalent. This heat sink has a black anodize finish. It  measures 5.5 inches long by 4.75 inches wide by 2.63 inches high and has  a thermal resistance of 0.67 degrees C\/W. It is listed by Wakefield  under their reference number 1025 extrusion. The Newark Electronics part  number for the 423K heat sink is 58F511. It can be found on their web  site. These heat sinks must be grounded to the central chassis ground.  The black anodize finish is an insulating layer. I recommend sanding it  off around one of the mounting holes, installing a solder lug under the  screw for that hole, and soldering a wire to central ground. Use an  ohmmeter to verify that the heat sinks connect to central ground. If you  don&#8217;t want to drill the transistor holes in the heat sinks, the  Wakefield 435AAAA is drilled for 4 TO-3 devices. It is a larger and more  efficient heat sink than the 423K. I do not know what the clearance is  between the transistors, so it may be impossible to drill the holes for  the bias diodes in the center.<\/li><li>Sockets for Q18 through Q21 &#8212; DigiKey 4601K-ND, DigiKey 4513K-ND  (Keystone 4513), or equivalent. The 4601K-ND sockets are sort of cheap  and I recommend them only if you cannot find the better ones. They use  No. 6 sheet metal screws to secure the sockets. Do not use machine  screws with these sockets for they will not hold! The 4513K-ND are a  little more expensive. They require #6-32 machine screws to secure the  sockets. I have had some students who have had these sockets crack when  the screws are tightened. Each screw should be alternately tightened a  little until both are tight.<\/li><li>Insulating wafers must be installed between Q18 through Q21 and the  heat sinks. I prefer the rubber type insulators which do not require  heat sink compound (DigiKey\/Bergquist BER100-ND). If you use mica  insulators (DigiKey 4662K), they must be coated with heat sink compound.  This stuff is very messy and will spread all over everything it comes  in contact with. I once heard someone say that if at dot of it is placed  on a wall near the floor, it will spread to the ceiling by the next  day.<\/li><li>Input jacks are panel mount phono jacks (DigiKey 576K) which must be  insulated from the chassis panel with mating shoulder washers. If the  jacks rotate after they are tightened, the shoulders are too thick. In  this case, use one shoulder washer and one flat washer on each jack. I  prefer the Switchcraft brand of phono jacks that have the teflon  insulating material. They are a little more expensive, but well worth  the price. If you use the Switchcraft jacks, you must use one shoulder  washer and one flat washer to insulate each from the chassis panel. The  part numbers for these are S1028 and S1029. In addition, you need one  flat solder lug washer for the ground connection to the jack. The lug is  a Switchcraft accessory. The Switchcraft jacks are available from  Allied Electronics.<\/li><li>Output jacks are 5-way binding posts (DigiKey J164 for red and J165  for black). Standard spacing between red and black jacks is 3\/4 inch.  You can also buy a dual 5-way binding post set, one for each channel. I  like them better than the single ones, but they are more expensive. If  you use them, tighten the nuts carefully. I once had one crack when I  tightened the nuts. Install R50 and C25 on the output jacks. Connect the  red jack to the circuit board loudspeaker output and the black jack to  the central power supply ground.<\/li><li>The Antek chassis seem to be a good value. Their eBay store is <a href=\"http:\/\/stores.ebay.com\/Antek-inc__W0QQ_fsubZ5\">here<\/a>. I believe that this company will make custom designed chassis. You download their software and design it. Then they make it.<\/li><li>Here are some suggestions for the chassis: <a href=\"http:\/\/www.budind.com\/\">Bud Valuline<\/a> model NHC-14156 or model HC-14103, Sescom model 3RU10, <a href=\"http:\/\/www.hammondmfg.com\">Hammond<\/a> model RMCV190513BK1, Lansing model BF10-V02B, and smarthome.com  model 873533. I have not seen any of these enclosures, but they look  like they will work. My model numbers should be checked to be sure that I  have the correct size. The enclosure must be at least 5.25 inches high,  by 17 inches wide, by 9 inches deep.<\/li><li><strong>Sescom Chassis Warning<\/strong> &#8212; On an amp built by a student who  used this chassis, we had problems because the main heat sinks were not  grounded. Not only does the rear panel have an insulating coating on it,  but it appears to be electrically isolated from the rest of the  chassis. If you use this chassis, sand the black anodize insulating  layer from around one of the mounting screw holes on each heat sink. Use  a tooth lockwasher under the screw head that holds the heat sink to the  rear panel and a tooth solder lug under the nut inside the back panel.  Solder a wire from the lug to central ground. You should verify with an  ohmmeter that the heat sinks are grounded. You should also sand the area  around the mounting hole for the central ground solder lugs to be sure  they make electrical contact to the chassis bottom panel. The bottom  panel also has some coating on it that acts as an insulator.<\/li><li>Power cord. I recommend the Digi-Key Q106-ND. The color code is as  follows: ac hot &#8212; brown, ac neutral &#8212; blue, safety ground &#8212; green.<\/li><li>Some students mount their heat sinks inside the chassis and use a  fan to circulate the air. A fan that one student found to be compact and  quiet is the radioshack.com model 9002516. It measures 3.1 inches by  3.1 inches by 1.5 inches and its noise is rated at 26 dB SPL. He mounted  the two heat sinks inside the chassis on the back panel and used a  circular hole saw to cut holes in both side panels. He mounted the fan  inside the box over one hole and a fan grill on the outside of the box  over both holes. The chassis depth must be at least 12 inches to do  this.<\/li><\/ul>\n\n\n\n<hr class=\"wp-block-separator\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">Some Recommended Parts Suppliers.<\/h2>\n\n\n\n<ul class=\"wp-block-list\"><li><a href=\"http:\/\/www.radioshack.com\">Radio Shack<\/a>. This is the  former Tech America, which changed its name to RadioShack.com. It is the  parts division of Radio Shack, and is not connected with the retail  stores. They have all transistors and many other parts. In the spring of  2002, students told me that RadioShack.com was selling its inventory at  steep discounts and would soon be out of business.<\/li><li><a href=\"http:\/\/www.digikey.com\">DigiKey Corp.<\/a>, P.O. Box 677,  Thief River Falls, MN 56701 (1-800-344-4539). Good source for resistors,  capacitors, some transistors, and chassis hardware. Competitive  pricing. This is a good company to buy from. I recommend all parts from  DigiKey for which their parts numbers are listed.<\/li><li><a href=\"http:\/\/www.jameco.com\">Jameco Electronics<\/a>, 1355 Shoreway Rd., Belmont, CA 94002 (1-415-592-8097). Similar to Digi-Key but less assortment.<\/li><li><a href=\"http:\/\/www.mcmelectronics.com\/\">MCM Electronics Inc.<\/a>,  650 Congress Park Dr., Centerville, OH 45459-4000 (1-513-434-0031).  Source of transistors. They also have some interesting looking rack  mount chassis.<\/li><li><a href=\"http:\/\/www.mouser.com\">Mouser Electronics<\/a>, 958 North  Main St., Mansfield, TX 76063, (1-817-483-0165). Source of transistors. I  have been told they also have the mica capacitors.<\/li><li><a href=\"http:\/\/www.newark.com\">Newark Electronics<\/a>, 4801 N. Ravenswood Ave, Chicago, IL 60640-4409 (1-800-4-NEWARK). Industrial supplier.<\/li><li><a href=\"http:\/\/www.apexjr.com\">Apex Jr.<\/a> They may have good bargains on filter capacitors and other things.<\/li><\/ul>\n\n\n\n<p>Newark and Allied have regional sales offices in many metropolitan \nareas. There are many surplus electronics companies which advertise in \nelectronics hobby magazines. The catalogs from as many such companies as\n possible should be obtained before acquiring parts for the amplifier. \nSurplus parts can cost as little as one-tenth the retail prices charged \nby companies such as Newark or Allied.<\/p>\n\n\n\n<hr class=\"wp-block-separator\"\/>\n\n\n\n<h2 class=\"wp-block-heading\">Power Supply for Two Channels<\/h2>\n\n\n\n<ul class=\"wp-block-list\"><li>T1 &#8212; The power supply voltages should be in the range (+ and -) 56 V  to 59 V with no load on the amplifier. There are several transformers  that can be used. These are as follows:<br> <ul><li><a href=\"http:\/\/www.antekinc.com\/\">AnTek AN-5440 500VA<\/a>. At late  2008, this transformer is priced at $57 plus $10 for shipping. It has  two 120V primary windings that can be wired in parallel and two 40V  secondary windings that can be wired in series. I have not tried one of  these transformers, but it looks like it will work. If the primaries are  not wired correctly, the fluxes will cancel and the transformer will be  a short circuit on the AC power line, and you don&#8217;t want that to  happen. It looks to me that the two reds on the primary should be wired  together and the blacks should be wired together. If the secondaries are  not wired correctly, the voltages can cancel. It is not clear how to  connect the secondaries, but I would guess that one blue should be wired  to one green to form the center tap. Use an AC voltmeter across the two  non-center tap leads to measure the AC output voltage. It should be  close to 80V. I use a fused Variac transformer when verifying the wiring  on new transformers, but not everyone has one of these.<\/li><li><a href=\"http:\/\/www.avellindberg.com\">Avel Lindberg Y236804<\/a>  500VA, 40VAC\/40VAC at $80. Avel Lindberg Inc., 47 South End Plaza, New  Milford, CT 06776, 860-355-4711. Mastercard and VISA accepted.<\/li><li>Victoria Magnetics  80VCT@6A toroidal transformer (model 05090001) for $62 ($58 with a  Georgia Tech student ID) or 80VCT@8A toroidal transformer for $82. These  are made by John Snowden <a href=\"mailto:johnbob@bigchair.com\"><em>johnbob@bigchair.com<\/em><\/a>  at Victoria Magnetics, 5945 Peachtree Corners East, Norcross, GA 30071.  The phone number is 770-448-8794. John can also make the E-I core  transformers. I do not have the pricing information on these, but the  cost should be less. The company John is affiliated with was in the  process of moving from Norcross to north Georgia during the fall and  winter of 2006. This could make it difficult to contact him. I do not  know the status of his move as of 12\/01\/106.<\/li><li>Signal Model 80-6 (80 VCT@6A, use 115 V primary taps) or Model 80-4 (80VCT@4A), <a href=\"http:\/\/www.signaltransformer.com\/\">Signal Transformer<\/a>,  500 Bayview Ave., Inwood, NY 11096 (1-516-239-7208). I have tested  these. Many students used them in the earlier days of the amplifier.  With an AC line voltage of 120 V rms, this transformer has a no-load AC  secondary voltage of just under 87 V rms. The DC power supply voltages  are + and &#8212; 58 V. The transformer description can be read <a href=\"http:\/\/www.signaltransformer.com\/signal\/products\/pdfs\/pgs24_26.pdf\">here<\/a>.  I was told in June 2000 that Signal wants $120 plus $40 setup charges  for the 6 A version. With prices like this, I don&#8217;t see how Signal can  sell many transformers.<\/li><li>EWC Model RPT 80-6 (80VCT@6A) or Model RPT 80-4 (80VCT@4A) split-bobbin rectifier power transformers, EWC Inc.,  385 Highway 33, Englishtown, NJ 07726 (1-732-446-3110). I have not  tested these. They appear to be similar to the Signal transformers. I  was told in June 2000 that EWC wants $120 for the 6 A version. Like  Signal, the EWC price is ridiculous, but unlike Signal, they don&#8217;t want  the setup charge.<\/li><li>Plitron Model 08A026000 (formerly ILP Model 8A026) (80VCT@6.25A) or Model 07A026000 (formerly ILP Model 7A026) (80VCT@3.75A). <a href=\"http:\/\/www.plitron.com\/\">Plitron Manufacturing<\/a>,  601 Magnetic Drive, Toronto, Ontario, Canada M3J 3J2 (1-416-667-9914).  These are toroidal transformers. My lab amp has the ILP 8A026 in it.  With an AC line voltage of 120 V rms, this transformer has a no-load  secondary AC voltage of 85 V rms. The power supply votages are just over  + and &#8212; 57 V. Plitron is the manufacturer and does not typically stock  its own items. They will accept VISA orders with a delivery date of  about 3 weeks. In early 1997, the cost of the 08A026000 was $76  (Canadian) each in quantities of 2.<\/li><li>Avel Model D4053 (80VCT@4.13A) or Model D4062 80VCT@6.63A), Avel Transformers,  Inc., 47 South End Plaza, New Milford, CT 06776 (1-203-355-4711). I  have not tested these. From some data provided by a correspondant, these  transformers seem to have the same voltage rating as the ILP  transformer I used in my lab amp.<\/li><li>Toroid Model 738.382 (76VCT@5A) or Model 749.382 (76VCT@6A), <a href=\"http:\/\/www.toroid.com\">Toroid Corp. of Maryland<\/a>, 608 Naylor Mill Rd., Salisbury, MD 21801-9627 (1-800-274-5793). I have not tested these.<\/li><\/ul><\/li><li>BR1 &#8212; 200 V 25 A Bridge Rectifier (DigiKey 26MB20A-ND)<\/li><li>C1P, C2P &#8212; Electrolytic capacitors. The value of the capacitors  should be about 8000 uF or greater. The dc voltage rating should be  several volts greater than the power supply voltage or greater. The  Mallory CGS123U075FF1, 12000 uFd, 75 V capacitor is a good example.<\/li><li>C3P &#8212; 0.1 uF 250 V capacitor (DigiKey E2104-ND). This capacitor  suppresses RF generated by switching transients in the bridge rectifier.<\/li><li>S1 &#8212; 15 A or greater AC line switch. Warning! The rocker type  switches that I have seen tend to stick. A toggle switch that makes a  noticeable mechanical click when switched is recommended. The smoother  sounding switches tend to stick.<\/li><li>PL &#8212; If a pilot light is desired, it should be wired in parallel  with the transformer primary. A neon light with internal dropping  resistor is recommended.<\/li><li>Standoffs &#8212; 3\/4 inch threaded standoffs should be used to mount the  circuit boards (DigiKey 1451EK-ND, threaded for 4-40 screws).<\/li><li>Fuses. F1 &#8212; 8 A slow blow. F2 through F5 &#8212; 5 A fast blow. Fuse  holders that mount inside the chassis are recommended (DigiKey 3536K-ND  for F1 and 3539K-ND for F2 through F5, the latter is a quad fuse  holder).<\/li><li>I have seen some amplifiers make the speakers thump on turn-on. This  is caused by the filter capacitors not being fully discharged. If  desired, bleeder resistors can be wired in parallel with C1P and C2P to  ensure that the they discharge completely. I recommend 3.9 kohm 2 watt  resistors for this. The power dissipated in each resistor will be  slightly less than 1 W. The resistors can be soldered to solder lugs  attached to the terminals of C1P and C2P. If the discharge takes too  long, you can go to smaller value resistors, but the power ratings must  be increased. The power dissipated is the power supply voltage squared  divided by the resistance. To be on the safe side, I like to have a  power rating at least twice the power dissipated in a resistor.<\/li><\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">Screw Hole Sizes<\/h3>\n\n\n\n<p>So many students have asked me what the correct hole sizes are for \nmachine screws. For a clearance fit for a number 4 screw, use a number \n31 drill bit (0.12 inch diameter). For a clearance fit for a number 6 \nscrew, use a number 27 drill bit (0.144 inch diameter).<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Version 4.4\/4.5 of the Amplifier Transistors (Starred types are preferred.) I tend to prefer Motorola if available. Motorola has spun off its semiconductor division, and it is now called On Semiconductor. In the process, they have discontinued many products. I have seen some mystery brand transistors that have really bad characteristics. If the transistor is&hellip; <\/p>\n<p><a class=\"moretag\" href=\"https:\/\/tomorrow82.ru\/?page_id=441\">\u041f\u0440\u043e\u0447\u0438\u0442\u0430\u0442\u044c \u0441\u0442\u0430\u0442\u044c\u044e \u0446\u0435\u043b\u0438\u043a\u043e\u043c<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-441","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/pages\/441","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=441"}],"version-history":[{"count":53,"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/pages\/441\/revisions"}],"predecessor-version":[{"id":494,"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=\/wp\/v2\/pages\/441\/revisions\/494"}],"wp:attachment":[{"href":"https:\/\/tomorrow82.ru\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=441"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}