Chip Amp Lm3886 Manual

Chip Amp Lm3886 Manual

The reason for its popularity is due to its very low distortion, minimal external components, and low cost. With the right layout and component selection, you can build an excellent sounding Hi-Fi audio amplifier that will rival high-end amps retailing for several thousand dollars or more. I’ll explain what each part of the circuit does, and show you how to calculate the right component values with examples from the amplifier I’m building. I’ll also show you how to layout the PCB and wire the amplifier in an enclosure for minimal noise and hum. I’ve also included the schematic and Gerber files for the power supply I used. It has all of the performance specifications, absolute maximum ratings, schematics, and design tips: It also has schematics for bridged and parallel amplifier circuits: This is an Excel spreadsheet that calculates output power, heat sink size, gain, and other useful parameters: At the end I wire up the amp so you can hear what it sounds like: Output power is what you’ll typically see stated as the Wattage rating of an amplifier. The maximum output power of the LM3886 is 68 Watts, but the actual power you get will depend on your power supply voltage and speaker impedance. You should be able to find your speaker’s impedance on the back of the speaker or in the user manual. This is the output voltage of the audio source you’ll be amplifying. It may be in the device’s user manual, but if not, you can get a rough estimate by playing a 60 Hz pure sine wave (there are apps that will do this) at full volume and measuring the AC voltage between ground and the left or right channel with a multi-meter. These calculations will tell you the correct voltage and VA ratings for the transformer you’ll use to power your amp. This step is important because if the transformer’s voltage is too low, the output power of the amplifier will be less than what you expected.http://www.etre-cheval.fr/Applications/MAMP/htdocs/etre cheval/news_pix/gouging-machine-manual.xml

• chip amp lm3886 manual, chip amp lm3886 manual pdf, chip amp lm3886 manual download, chip amp lm3886 manual free, chip amp lm3886 manual 2017.

If the transformer’s VA rating is too small, the amplifier might clip or distort the audio at higher volumes. Your amplifier’s peak output voltage will depend on your desired output power (P o ) and speaker impedance, according to this formula: This is the voltage the amplifier needs from the power supply to get the desired output power. Then factor in your transformer’s regulation and the variation in your mains voltage. Regulation values can usually be found in the transformer’s datasheet, but if you don’t know your transformer’s regulation, a safe value to use is 15%. The regulation of the transformer I’ll be using is 6%. It usually peaks late at night when people are asleep and drops in the daytime when more people are awake and drawing current from the power grid. The DC voltage will be higher after the bridge rectifier diodes on your power supply convert the AC voltage to DC. With an 18 V transformer, I would get a maximum supply voltage of: Total power depends on the maximum supply voltage output by the power supply, the amplifier’s peak output voltage, and the speaker impedance. The formula to use is: Sound quality is severely compromised when the SPiKe circuitry is enabled, so to prevent this we need a heat sink with a thermal resistance low enough to dissipate the maximum power dissipated by the LM3886. P dmax depends on the maximum supply voltage of your power supply and your speaker impedance: Heat sinks with a lower.The LM3886T has a metal flange on the back of the case, and the LM3886TF is all plastic. The plastic case of the LM3886TF gives it a higher ? cs: I’ll be using the schematic below.http://www.blackbookmedia.co.za/uploads/gould-gt15-manual.xml

It’s basically the same as the one in the datasheet, but with the optional stability components included: The minimum gain setting of your amplifier will depend on your input voltage, speaker impedance, and output power according to the formula: These resistors form a voltage divider that determines the voltage at the inverting input (pin 9) of the LM3886: Setting it too low might make your amplifier too quiet. A good gain setting that’s not too high to cause distortion, but not too low to give you a good range of volume is around 27 to 30 db. If these resistors vary much between the two channels, the gains will be different and one channel will be louder than the other. Metal film resistors with a tolerance of 0.1% or less are ideal. This difference in voltage will be amplified as noise. You already found a value for R f1 when you set the gain of the amplifier. The values of R in and R b are chosen so together they equal the value of R f1. This will make the current at the non-inverting input equal to the current at the inverting input. To find values of R in and R b for a particular R f1, use this formula: It’s main function is to block any DC present in the audio source, while allowing AC (the audio signal) to pass. DC in the audio source needs to be blocked or it will be amplified along with the audio signal and create a high DC offset at the speakers. This distorts the audio, which we don’t want for obvious reasons. In a high pass filter, frequencies below the cutoff frequency will be attenuated (muted). In a low pass filter, all frequencies above the cutoff frequency will be muted. We’ll use combinations of low pass and high pass filters to set the amplifier’s bandwidth and improve stability. The lower limit of human hearing is 20 Hz, so the F c should be well below that to prevent attenuation of bass frequencies. Lower than 2 to 4 Hz is ideal. I started with 1.5 Hz, but you can use higher or lower values if you want.

Just be sure to stay well below 20 Hz or the bass will be weak. Electrolytics, ceramics, and tantalum capacitors should be avoided. A good quality polypropylene metal film, or even better a polypropylene metal film in oil capacitor will sound best here. If the F c of this filter is higher than the input filter, the amplifier will pass low frequencies to the feedback loop that it can’t handle. This will create a voltage across C i and cause DC voltage to appear at the inverting input, which will be amplified and cause distortion. Therefore, the input filter (C in and R in ) should determine the lower bandwidth frequency of the amplifier, not the feedback loop filter (C i and R i ). With smaller values of C i, the bass will be softer and have less punch, but with larger values of C i, the bass will be tighter and have more impact. We can use the F c equation with R i and C i: The capacitance will probably be too high to use polypropylene, so you’ll likely have to use an electrolytic. However, there are good quality audio grade electrolytics like the Elna Silmic II or Nichicon KZ series which shouldn’t adversely affect the sound quality. In that configuration, C c filters radio frequency and electromagnetic interference picked up by the input wires. Unfortunately, it also increases the chance for oscillation. A better way is to connect C c from the non-inverting input to ground as shown in the image above. That way C c still filters radio frequencies, but it also acts as a low pass filter that will set the upper limit of the amplifier’s bandwidth. Broadcast radio frequencies in the USA are: It’s well below the lowest AM broadcast frequency (535 kHz), so radio frequencies and most electromagnetic interference should be filtered out. It’s also well above the upper 20 kHz frequency of human hearing, so higher audio frequencies won’t be attenuated. With a 680 pF capacitor, the F c becomes: C c is also in the signal path, so a good quality capacitor should be used.

The best dielectric types for audio capacitors in the picofarad range are silver mica or polystyrene. That’s beyond the scope of this article though, so I’m just going to use the values given in the datasheet. It also prevents radio frequencies picked up by the speaker wires from getting back into the amplifier’s inverting input via the feedback loop. R sn limits the high frequency current so there isn’t a direct short to ground, which could exceed the current limit of the LM3886. Therefore, smaller values of R sn make the Zobel network more efficient at filtering radio frequencies, but it also increases the cutoff frequency, which in turn reduces it’s effectiveness. This makes the F c: In order to bring this down to a more reasonable level, I decided on using 4.7 ? for R sn and 220 nF for C sn, which lowers the F c down to 154 kHz: C sn should have low ESR and low ESL, with a voltage rating greater than the rail to rail swing of output voltage. To minimize inductance, locate the Zobel network close to the output pin (pin 4) and keep the traces short. It also prevents radio frequencies picked up by the speaker wires from getting back into the amplifier’s inverting input through the feedback loop. Audio signals are relatively low frequency, so they will flow through the inductor uninhibited. High frequency oscillation current will be impeded by the inductor and be forced to flow through the resistor, which will dampen it. They should be located away from the amplifier’s input circuitry to prevent interference from the magnetic fields generated by the inductor.Since the inductor will be carrying the full output current of the amplifier, the wire should be heavy gauge. 12 to 18 AWG would be good. Use this Single-Layer Air Coil Calculator to find out how many turns you need for a particular wire diameter and coil diameter. The diameter of 14 AWG is 1.62814 mm. I planned on using a screwdriver shaft with a diameter of 11 mm to form the coil.

The negative supply pin needs it’s own set of decoupling capacitors and the positive supply pins share a separate set of decoupling capacitors. Larger values will improve the bass response.They also filter noise and radio frequency interference in the power supply. Inductance and resistance inhibit the flow of current, which increases with longer wires and traces. Since the power supply is relatively far away from the chip, inductance and resistance are a problem. To maximize current flow to the chip, the decoupling capacitors should be placed as close as possible to the chip’s power pins. That is what I’ll be using in my amplifier. The 16 V Zener diode (D1) blocks current flowing out of pin 8 until the diode’s breakdown voltage (16 V) has been reached.If you don’t want do all of the calculations we did above, you can use the values I used. Here’s the final schematic: Click on the image to edit the schematic or change component values. With a properly designed grounding layout, the amplifier’s output will be completely silent when the source is connected and no music is playing. With a poorly designed ground layout, the amplifier can produce a very noticeable hum or buzzing sound. Low current grounds are the ground feeds to the input circuitry and feedback loop. High current grounds are the ground feeds to the power supply decoupling capacitors, the Zobel network, and the speakers. High currents flowing through the low current ground conductors will create a DC voltage that can show up at the amplifier’s input and get amplified as noise. The main system ground is located as close as possible the reservoir capacitors on the power supply. The main system ground will connect to the mains earth wire via a ground loop protection circuit (explained later), and the amplifier chassis. The diagram below shows how to order the ground connections: Below I’ll discuss the guidelines I used to design this PCB layout.

The PCB is for a single channel, so for a stereo amplifier, you’ll need to build two boards: You can click on the image above to edit the PCB layout, change the component footprints, and order PCB’s. You’ll be able to choose the copper thickness, PCB thickness, color, and order quantity: All of the traces and printing came out very clean and precise, and there were no defects on any of the boards. Here’s one of the PCBs: Placing their PCB terminals on opposite sides of the board will make it easier to keep them separate when you route the traces. The traces for the power supply feeds and power ground should be routed close together to reduce the loop area. Likewise, the audio input and signal traces should be routed close to each other. An easy way to minimize the loop area is to use ground planes on the bottom layer of the PCB, which I’ve done in this layout. Each one has its own electrically isolated ground plane on the bottom layer. Since the power ground carries high currents and the signal ground carries low currents, they’re kept separate until they connect at the main system ground. On the top layer of the PCB, the power supply, output, and Zobel network traces are routed over the power ground plane. The input and feedback loop traces are routed over the signal ground plane. The traces for the power supply feeds were made very wide to minimize the resistance and inductance. I trimmed the leads of the feedback resistor (R f1 ) and soldered it directly to pins 9 and 3 to keep the loop area as small as possible: Since inductance increases with trace length, it’s best to keep all traces as short as possible. This is especially important for the power supply decoupling capacitors, feedback loop, input circuitry, and Zobel network. Keep the components for these circuits right up against the chip’s pins so the traces will be short. This will give you a stronger solder joint and better electrical conductivity.

Start soldering the smallest components first, and work your way up to the larger components. If the component moves in the pasty phase, it can create a cold solder joint. The smaller melting range of eutectic solder makes the solder set faster and gives a better electrical connection. Metal enclosures are the best type because they shield the amplifier from interference caused by fluorescent lights, radios, and cell phones. Unfortunately it can be hard to find a chassis that fits everything and looks nice too. After a lot of searching, I found a company called Hi-Fi 2000 that manufactures some really nice metal enclosures. Their website is in Italian, but it can be translated to English.Then measure the overall dimensions to make sure the enclosure will fit everything. The wiring layout is just as important as the PCB layout and grounding layout. Use the diagram below as a guide for wiring the various parts together: The audio input wires and signal ground wires are the most sensitive to interference from surrounding magnetic fields. If you orient the input side of the amplifier PCBs near the input terminals on the chassis, the wires can be kept short and away from sources of interference. To minimize the loop area, the following sets of wires should be twisted together tightly: I used 14 AWG, but anything larger than 18 AWG should be fine. I used solid core 22 AWG, which works well because it can be twisted into a tight coil. If this becomes a problem, you can install a 1 nF capacitor between each input terminal ground and the chassis to filter it. I’d also use a lock nut or lock washer to prevent it from getting loose. All metal parts of the amplifier (like the heat sinks) should be electrically connected to the chassis to provide a path to earth for any mains voltages that could contact them in the event of a fault.

The ground protection circuit can connect to the chassis at the bolt where the mains earth wire is connected to the chassis, or at a separate location. The right channel PCB is mounted upside down so that the input side of the board is close to the RCA and 3.5 mm input terminals. In this arrangement, the heat sinks provide some shielding from the Thiele networks and the AC wires leading to the transformer: This is know as a ground loop, and it can create hum in your amplifier’s output. The resistor reduces this current and breaks the ground loop. If a high current fault occurs, the fault current can flow through the diode bridge to earth. Note that the chassis MUST be electrically connected to the mains earth wire to prevent mains voltages on the metal chassis in the event of a fault. The capacitor is there to filter any radio frequencies picked up by the chassis. The “PSU 0V” terminal connects to the main system ground. The “Chassis” terminal connects to the chassis: It’s the best amp I’ve ever owned by far. The bass is very deep and clean. You can really feel it. The highs are clear, but not harsh at all. I can hear details in songs that I never knew were there. Trust me, if you build an amp with the LM3886 you will not be disappointed. It definitely lives up to it’s reputation as a Hi-Fi amplifier. The video at the beginning of the post will give you an idea of what it sounds like. Due to the length of this post, I decided not to cover the power supply in detail, but I may do so in the future. And be sure to like, share, and subscribe if you found this helpful.A single chip for each tweeter and a parallel pair for the woofers. This setup drives them quite well. I also really like how you supplied the calculations for determining the size of the components. As others have stated, I’d love to see a followup on the power supply. If not a full post cover it, could you at a minimum post a link to the pcb?

The LM3886 is an excellent component and you explained “who to do better”. Shouldn’t you also consider the voltage drop on the rectifier diodes, when you calculate the maximum voltage required by the amplifier? thanks. I think this is an omission in the article.Do you think you could post the exhaustive list, brandd and referenced of the components ? thanks. I second the request for component details, such as the voltages of the caps that can’t be read, if possible. I learn a lot, specially about grounding. Congratulations! A complete schematic, as well as components values for the power supply will be greatly appreciated, Thank you so much in advance. When you convert the transformer’s AC voltage to DC on the power supply, the voltage actually increases to about 25V due to the bridge rectifier. A lot of effort obviously went into this. Many thanks! Many thanks too for the fantastic links. Thus, I was wondering if you could state the specifications of your amp, (distortion and noise characteristics) so it can be compared to other LM3886 designs, so see if one needs to improve anything. I will definetly try a build on my own. Sascha Quick question, I’m making these amplifiers to work in my car, so I will be using a 12V-30V DC transformer to power it. Should I still include a ground loop protection circuit or is it not needed for this approach? The two silver mica capacitors (47pF and 680pF) appear to be reversed in your pictures.As C13 is used to stabilize the amplifier I would just like confirmation that the parts list and silkscreen are correct. Thank you. The 680 is in fact the smaller-, the 47pF is the bigger one. Success with the build, Ed Did you forget it? I have a transformer that feeds 22V at 7A. I can use it, no? Good luck, Ed I hope you’ll make more of those overly detailed explanations, amazing knowledge source. I would not call this basics. There are many deep understandings beneath the guide. I like the way you treat ground connections.

But I think three wire transformer output should also do just fine. I would like a kit to same the time to order them. I do not care particular brands of capacitors or any other components as long as their specs meet requirements. Firstly, if we want a volume control for this amplifier, how we will achieve it. Second is there any need for preamp as my audio source would be a laptop or a smatphone. Maybe it helps other people with similar questions. Please double check the information below, as I do not have alot of knowledge about electronics yet (I am a newbie). You can add an input selector knob (if you want more inputs) on the same controls-pcb and then mount the pcb parallel to where your control knobs should be. Most of the time the controls pcb will be mounted parallel towards the front of the amplifier. Most designs have extensionrods are used between the pcb and the controlknobs at the front, to obtain a better internal layout. You do not want wires going from the back to the front of the amplifier and back again, so power components and pcb’s in the front half of the amplifier and signal pcb’s and components in the back is optimal imho. My assumption is that how more “stuff” you add between the signal-input and amplified output circuit, the more distortion you could get (seems logical). If you want to add controls to your amplifier, design your controlscircuit of similar high quality as the (munford) components and circuit designs used. Like, I am going to try to design a controlscircuit for this diy amplifier project, that I can add later on in the case as an upgrade, before I start building this diy amplifier project for real. However I am almost a new graduate of Electrical Engineer and not sure if I can handle the processes. Anyone has any idea on how to be qualified in order to build this amplifier without major mistakes. Thank you in advance. I really enjoyed reading this awesome guide.

I would like to know if there is a way to buy the power supply PCB instead of making it from scratch. Thank you. I don’t have the patience to start this project can you advise me on where to find a really good lm3886 assembled board.The sound is fantastic, however I can not reach it but the power is far from expected. What could be wrong. Can you help please. Thank you. Nheco I think you’ve completed the most comprehensive and understandable post on this subject. I can’t thank you enough. And I have a few questions: Or is it just because the part is only manufactured like that by Mundorf? I was thinking of using one of the smaller WIMA MKP 4 series capacitors for C1. Those are also metalized polypropylene film capacitors, but with a smaller footprint and lower voltage ratings.All other (47) pages of the article and the comments are hidden. In there you could put things like a publishing date and a copyright notice, keywords, a table of contents, (social link) URLs written in full as normal text instead of just blue words in the article, etc. From then on, I have been trying to improve its performance by making modifications. In addition, I wanted to publish a systematic article with instructions on that. However, due to lack of deep knowledge in electronics, the project was lagging behind. This is amazing. Though I have no expertise in Physics, was able to understand the technical things in this article. Awesome job, guys. I noticed that in the schematic you have a 100nF cap (Csn). And in the photo under Wiring It All Together, there is a polypropylene cap labelled as a 100nF cap (I assume this is Csn). You explain in the tutorial (and as the actual photo of the cap shows) the actual value of the polypropylene is.22uF. Could you update this info. Thanks. This is my first and through the detailed explanation, I’m well able to understand almost all of it. I have a question though.

For the power supply circuit, how can I connect diodes and capacitors to safely give the required voltage and do away with the transformer. Thanks in advance. Rb before Rin or Rin before Rb. Cin after Rin? Cc before Rb? Cc before or after Cin.Use low ESR bulk storage caps and minimize inductance lengths. Keep all unregulated storage caps close to the LM3886 with proximity based on parasitic inductance effects. Meaning, small caps closest to LM3886. The combination of bulk caps should provide low impedance current throughout the frequency band for the load. Also, at the same time provide IC centric decoupling of the supply. However If amp is run at near full load (not likely) heating effects of the transformer and bulk storage caps should be cooled as well. (This is where you find air flow holes across the entire top and bottom of enclosure.) Use the metal tab version of the part. Published Rth of plastic is optimistic. The design should specify based on cap heating RMS Ripple current effects (ESR), which basically translate into Capacitor life expectancy. So depending on the capacitor supplier used, the design may have larger bulk capacitance, divided into multiple smaller caps in parallel as needed that can satisfy the RMS ripple requirement. Also, some airflow spacing is needed around the caps. A person with or without electronics knowledge can follow this quite easily. This helped me have the courage to continue with my own Gainclone amp project. THANK YOU. I just finished my build based on your tutorial, with minor changes. At least to me, but I’ve been listening only to Yamahas for the past 10 plus years. Now after a couple of hours of listening to the LM3886 I must say it really sounds great. I think the sound my LM3886 amp produces is much sweeter and softer than my Yamaha. But I was astounded. I listened to songs I heard hundreds of times with a slack jaw. Amazing!

Also, due to the sweeter and softer nature of the sound I’m able to comfortably listen to it on a volume which I would hardly stand listening to on the Yamaha. Any idea what it can be? Get from a reputable supplier (components specialists rather than ebay). I sure hope I got genuine chips as I ordered them from Mouser. Funny thing is: my LM3886 amplifiers from AliExpress and eBay are both working fine. Pretty ironic. I want to use PC’s intagrated Realtek sound card as signal source. Which voltage level gives this output. Probably 1 volt too? It just does not get better than that. However, is it possible to go directly to EDA with info you posted for the amplifier board. Do they have any kind of records on that so it may only require a push of a button on this end? Robert You use the TF version, why is that? Thanks. The TF version is isolated and mounts directly to the heat sink. This allows for a smaller heat sink for the required dissipation thus saving a considerable amount of space and cost. Robert You explained it very well to me. All the best John It is hard to resist from building this amp:-) With two two separate power supply units, all grounds be channel specific on the PCBs and the two ground-loop protection circuits (one from each power supply unit PCB) connected together to the chassis? I have just a question: in section “SET THE LOW FREQUENCY CUTOFF AT THE FEEDBACK LOOP”, how did you obtain the formula for Ci. I would like to better understand how to set a value for low-frequency cutoff at the feedback loop. Thanks in advance. From my understanding the 50pf should be C13 and the 680pf the C2?? that had me confused for a while because of the pcb layout and the image. Anyone?

If I am correct the pcb layout needs to be modified to represent the difference in size between the 50pf and the 680pf My suggestion would be to not just follow the instructions, which are very well explained, but to take full advantage of the instructions as a starting point to a better understanding of exactly what each component is doing. The means to do so have been provided. Both the lm3875 and the lm4780 which are both very impressive. Soldering quality does make a noticeable difference, don’t bend pins at the body, will cause damage Military grade Silver Mica capacitors SSG2 0.1uf at 0.3% tolerance (possibly the best capacitor made) 615pf at 0.3% tolerance FT3 teflon 0.47uf and 0.22uf and 680pf Polystyrene 4uf, 1uf, 0.22uf and 680pf and 48pf All at amazing prices when compared to what they would cost here if obtainable without custom order. Every component really does make a noticeable difference choice is subjective for example; it seems people like the foil in oil because the oil allows movement which causes distortion in 2nd order harmonics, which people perceive as warm. Too much clarity is not something people are used to, which can be an issue still experimenting Crossed over at 1000 and 7000, that’s about it No matching of different drivers resonance frequency etc then crossover to match It is definitely worth building ones own speaker system using a dsp rather than a cross-over, same as the Gainclones one can have exceptional performance at a fraction of the cost if one is willing to put in the time and study I am now in the process of building 3 way active speakers, initially with crossovers then with dsp’s fed through a preamp. The advantages of the amps in the speakers are immense and save on cost of cases binding posts banana plugs length of wire etc as well as massive improvement in audio quality. I have quite a bit of leftover CPU coolers and i have an idea to use those for building an amp based on 3886.