Yes you can use gate drive transformers. I would advise to check the output of the transformers with an oscilloscope first before you apply power to the IGBTs. Check for ringing, proper voltage, proper rise times, and proper polarity with the transformers connected to the IGBT gates first before you apply power to the DC Bus. This will save your IGBTs from damage, or possibly from destruction. Thanks for watching
Technically speaking you can get away with using 3 isolated supplies. One for each of the upper IGBTs and One for both lower IGBTs. Notice on a full bridge diagram, where the emitters are located on the upper IGBTs; they are in common with the output phase. You can't use one supply for both uppers because you would be shorting out both phases with each other. The lower IGBT have their emitters in common with the DC Bus negative, so you can use one supply. Thanks for watching
so would the igbt modules work good for a DC to ac inverter. h bridge. and for the DC to dc boost . I'm kinda new at all this. but need to build a DC to ac inverter. to run my home. it just seem a few igbt modules would be simpler than a few dozen MOSFETs. I would like to make it where I can expand the potential of it. like 3kw. then double that 2-3x later as I need more power. I guess what I'm wondering is do they switch fast enough to be used in a spwm inverter. I think its a 23-24khz carrier and 60 hz sine wave. thanks and I enjoy your videos
Hey, For reverse input voltage protection,I decide to employ diodes in series with each IGBT,will an inductive load be able to discharge when the switches are off since the free wheeling diode will not be conducting?
Hi mate, love your stuff... I've learnt heaps from your easy-speak approach. I do have a question for you though and I'd love to see how you would explain this. I don't think you've covered this subject in any great detail as yet (correct me if I'm wrong). Could you please cover the effect of gate resistance on the dead time in inverters. ie: The delay caused by gate resistors and the dead time required in real circuits such as inverters. Ta mate.
what value gate resistors are you using also looks like you have a resistor across gate to emitter on the igbt whats the purpose (pull down ?) and what value .. cheers
Great video, very informative! Can I ask what type and size of capacitors you used for this circuit. Both for the capacitor bank on the igbt and the caps used with the 555 timer to get a ac signal. I'm a little confused about the sizing of them. Thanks again for the great video.
Hello!! Thanks for all you knowledge you post here. I want to know, if I am going to use an IGBT (CM400HA-24A) as a dc motor speed control from 30 A123 20ah cells must I use isolated dc for the gate driver??
So if u wanted to create a induction heater capable of melting metal (i wanna get away from propane). Would you say a full bridge inverter to drive the inductor is better than say a half bridge version (not really understanding the differcen). Been trying to read up on the subject.. but i make knifes.. not induction heaters. ;D
A half bridge circuit will only give half the output voltage. A full bridge circuit will give full output voltage. The output of a full bridge circuit will also be smoother, the currents and voltages will be more steady
Could you share your schematics. I would like to apply it for high power application. What frequency do you use and what is the dead time of the square wave generator? Thanks
Thanks a lot Man this is awesome! You just cleared my head of some nagging questions I've had for some time now.This might be a dumb question, but are those 555 timers on the power supply?
thank mate to share this video , but I have questions 1.what is the resonant frequency of the tank circuit 2.do you use any feedback control system in it?
Hey, One more thing, Will it cause "Short-Circuit" if I use Same Driver Power Supply for "Upper Two IGBT" & Other isolated Second Driver Power Supply for "Lower Two IGBTs" ?? 2 Isolated Supply is Enough ??
Hey Buddy..! May I use 4 different Transformers also instead of Battery for Driver Supply..( 12-0-12 Volt, 1 Ampere ) ? Anyways this Video is Superb .. Keep it Up.
if you could give me an idea frequency to switch my igbt (1khz - 18khz) and some of the benefits of higher vs lower frequency. I figured out most of the small things I'm trying to figure out how much heat this thing should produce. hearing the switching frequency doesn't bother me I have a currently set at 18 kilohertz but gets warm rather fast. I have a heat sink measurements 2 inches tall 7 inches wide 10 inches long, also I tried emailing you but I'm not sure if you got it or if I'm in your junk bin.
Try mailing me again, but put my youtube username in the subject line. That should help it get to my inbox. You've already figured out the most important part about switching frequency. Higher switching frequencies produce a smoother current output, but they also dramatically increase the switching losses. You can lower the switching frequency to reduce IGBT heating, however if the output current becomes too rough, you may need to add in passives (inductor, capacitor, etc....) to smooth the output current
one more thing the square wave output from the full-bridge has a frequency equal to the frequency of the PWM signal from 555 Timer. how do we convert this high-frequency signal into a 60Hz signal???
could this run at 60 hz. and 120v for a DC to AC power inverter. or as a DC to DC boost stage for a inverter. I am needing to build a inverter to power my off grid home. I am self taught. and learning all I can to build my inverter. for like 24vdc to 120vac. thanks for the videos. they have really helped. but I would like to know more about the brain of the system. what powers the FETs or igbt. I have thought about just using a ic555. but I don't think it will do what I need. I need to regulate output and compensate for battery voltage drop. so its the boost stage I'm in sure about. if you could give any help or advice. I will be grateful and thanks again!
Yes you can run something like this at 60 Hz, but that would output a square wave and that isn't what you want. You want something that is running PWM that simulates 60 Hz. The easiest way really, is to get yourself a big VFD and an output filter. Set the output frequency to 60 Hz. Then make a boost converter to feed the VFD DC bus input. Of Course, You can make your own inverter that will work as you want. Several micro controller chips are available that are pre programmed to do the job you have in mind. MC3PHAC is a motor controller chip that can send drive signals to gate drivers to switch IGBTs in the correct pattern, you just need to provide the capacitors, oscillator and resistors. Then it would be the same, just feed the DC bus with your boost converter.
+subcooledheatpump thanks for your reply. I still haven't got to far on my project. but I found a inverter driver board with get drivers. looks like it will work. and I was thinking of using a arduino for the DC boost stage. and was thinking of using a center tap push pull transformer. but to put out 6-8kw. it would be huge. and limited to low frequency. I was also of building a core from iron powder and epoxy. basically use a laminated iron core transformer as a model. or go with a toroid of a decent size. maybe 10-12" and see if it would run at 1khz or so. I would like for it to be oversized. so I can expand the inverters output later. to around 10kw or better. as my battery bank grows. I would love your advice. you seem to really know your stuff! and your videos are a very useful learning tool and I have watched a few multiple times. I have learned lots from them. thanks again
Hi bro i want to use semikron igbt skm200gb063d for induction furnace but i don't have SOA graph of that igbt i find in datasheet but that is not there how i can find max power of that igbt can you help pls....
Thank you for your speedy response!! is there a way I could get you to help me with some details?? I would pay you for your time and experience. If not its OK
hi dear i want build a single phase motor controller i made 2 square wave with SG3525AN ic that i can change the frequency of this two wave 9hz up to 300hz if i want drive 4 igbt with this square wave i have to use the hcpl ic? i can not understand what how can we connect the negative of 15 volt source for vcc of hcpl ic to emitter of igbt because we have the flow of current to load i mean this current dosent harmful for hcpl ic.thank you so much
I don't fully understand this about the fully isolated power supply, and isolated grounds, for instance if you have build a linear power supply, does that mean that you have to use 4 transformers, 4 regulators one for each IGBT driver?, or you can use 1 tranformer, 4 regulators? Why must Vcc for the octocoupler be even isolated, isnt' that just power supply for the IGBT. I'm confused. Thanks in advance.
hai !!! nice work ..one question though !! you speak about four ISOLATED power supplies..but on your diagram they all seems to connected by ground bus..please clarify on this..thanks in advance
Hey, very nice explain, thz gate driver using and why interverting. But, i f i want ro send just cycle lf swiching ( i want these 4 igbt switch off, how i can get that? Is needed to send all at gnd ?
To switch an IGBT off with a gate driver, all that's needed is removal of the input signal to the gate driver. Without a gate driver, simply short the emitter and gate terminals
Some gate drivers can drive both high and low side and provide a form of isolation that is built in. You can use just one power supply for the low side IGBTs, but the high sides are where the isolation is required.
you are using gate resistor at the output of driver ic.what is the value of second resistor that you have solded on IGBT module.how you have selected that resistor.i assume you have connected that resistor between gate and drain of IGBT.this resistor has not been shown in circuit diagram.is this resistor is to charge and discharge gate capacitance of IGBT module.please correct me.
The place I bought those IGBTs from sometimes would have a 1K resistor soldered between the gate and emitter, this keeps the gate from being damaged by static buildup during shipping. It's also sometimes helpful if you aren't using negative gate drive/bias, just to drain the gate and be sure it won't turn on without a signal from the gate driver
is there any way to calculate the value of resistor connected between gate and emitter by using data sheet.i have build the circuit of induction heater but one of my module is heating while second is normal.is this due to square signal of IGBT gate driver.can I trigger gate using sine wave.if yes then how to implement dead time technique. is it possible to reduce heating by reducing deadtime.i mean is heating effect is deadtime dependent.
hello i want to make three phase inverter for speed control of 3 phase 5 hp indution motor so please tell me whats rating of igbts should i use and also other components please..............
A full bridge "inverter"? Interesting! One way of classifying power conversion systems is according to whether the input and output are alternating current (AC) or direct current (DC), thus: DC to DC = DC-to-DC converter, Voltage regulator, Linear regulator AC to DC = Rectifier, power supply unit (PSU) DC to AC = Inverter AC to AC = Transformer, Voltage converter, Voltage regulator, Variable-frequency transformer
The IGBT turn off spike also has an effect on the bus voltage. The main DC bus capacitors won't response fast enough to quench the initial spike. So the metal film capacitors help with that.
nice dude. also check out this crazy chip. STGIPS30C60T-H . it includes drivers and IGBTs all in one, just needs PWM st right from MCU to chip. only thing needed is your shunt current measurements, the MCU and caps.
That usually isn't managed by the IGBTs or the gate drivers, but by the PWM generator which supplies the gate drivers. Many PWM IC's feature built in dead time control, MC3PHAC is one example of a PWM motor controller which features built in dead time control
Thanks for reply, Suppose I'm giving pulses to igbts using Microcontroller then it's necessary to give pulses with dead band difference? Means suppose I'm using h bridge mode at that time I'm switching on diagonally igbt pair and then other pair diagonally...At that time between these two switchings I should put dead band?
The Rg is very application dependent. IGBT manufactures have a list of generally acceptable Rg values for a given IGBT. In general, Rg must be raised if higher voltages are to be used, or if the turn off voltage spike is too high. Rg must be reduced if the thermal dissipation is exceeded, or high switching frequencies are to be used. As for the isolated power supplies, the high side (positive) IGBT gate drive power supplies MUST be isolated, unless you like explosions. The low side gate drive power supplies don't need to be, but some are anyway to reduce the possibility of noise or interference
subcooledheatpump Thanks mate for answer. That and some more things, I would only like to propose you, to show in one of your videos, there are not many videos out here, how to drive this IGBTs blocks, I mean the real "basics", showing also the use in practice, Maybe on some power managment too, like 'phase-shift control of H-bridge'(I picked that up from richie brunnet site*) and others.. I greatly appreciate..
I wouldn't recommend to solder part on IGBT brick. It can low lifetime or degrade IGBT by heat. Also it is unsafe. If spark, IGBT might died. Hundred dollar draining. Gate part are not very strong as you think. Spark can happening anytime.
+paul bergin For all the questions about dead time, Dead time is simply the required time when all IGBTs are switched off. For each IGBT, a value is specified in the datasheet for these characteristics; Turn on Delay, Rise time, Turn off Delay, and Fall time. You will notice that most IGBTs usually turn on faster than they turn off. In Inverter circuits, such as shown in this video, this could cause both high and low side IGBTs to be on at the same time, causing a short circuit known as "shoot-through" (IE. lower would switch on before the upper switches off) Dead Time simply turns off both IGBTs for a very short time (usually around 1 microsecond) to allow each IGBT to fully turn off before the opposing IGBT turns on. Gate resistance has more of an impact on turn on time, as the IGBT's gate is a capacitor, higher resistance leads to a slower turn on. This could help relieve some mismatch in turn on and turn off times. It is possible to use two different gate resistors on the same IGBT, one for turn on and a different one for turn off to match the rise and fall times, this helps reduce or eliminate dead time. The gate resistance does affect turn off, but only so much, as IGBTs like other bipolar transistors feature a parasitic "current tail" or fall time. This current tail will limit the IGBTs turn off speed, and is thus the reason inverter circuits need to use some dead time.
+subcooledheatpump Thanks heaps buddy, this is what I imagined was the case, the difference between the turn on and off times for my Ixys IGBT's is quite noticeable on the scope and would definitely lead to problems with shoot-through as you up the drive frequency. So in general a higher gate resistance will delay the on saturation time which will alleviate the risk of crossover time, I also understand that too much delay will lessen performance. I have set up a test rig to visualize the actual gap between output pulses on my scope whilst running at 20khz. I will aim for 1us. Ta mate.
What is this unit called for and what does it do? Found 2 of these with hot-swap(?) handle. 3 Pictures here: postimg.org/gallery/2gophpmlk/ Than serial number is for transistor module and that way I found your video.
I'm not understanding why you say each "gate driver" needs an isolated power supply. Both high-side drivers could share a supply, and both low-side drivers could share another... but why not use real gate drivers with charge pump... no worries at all.
Only the low side gate drivers could share a power supply. The high sides need some form of isolation between them, otherwise the bridge output will be short circuited
Therein lies the reason the full bridge charge pump gate drivers exist... like SM7295, A3921, DRV8704 (dual H-bridge), MC33883... and many more. Even the ooollld faithful IR2110 is great for this... it just takes two of them for a full H-bridge.
@@beforebefore Yeah charge pumps are ok for lower power applications. Personally I never had too much luck with them. Probably did something wrong I'm sure. Most manufactures recommend isolated power supplies, even for the low sides after a certain power level.
I like your gate drive. I would ask if you could shed some light on a FullB induction heater listed on instructables. www.instructables.com/id/30-kVA-Induction-Heater/ If you notice the isolated GDTr's are wound toroids. I have looked at this design and am a bit condused about the details of the windings. You sound like you could clarify that issue and have most likely used that system before. Can you please share anything? Thanks.
The gate drive transformer he used was toroidal, that just means the windings were wound onto a ring like transformer structure. They are electrically the same as any other transformer. As for how they work, there is one primary coil, driven by an amplifier stage. There are four secondaries. with two of positive polarity, and two negative polarity, that is, with one half switching cycle. On the other half of the switching cycle, they reverse. This is used to drive the opposing IGBTs on and off at the right time to avoid shoot through. Basically the same thing as you see in my video, just done with gate drive transformers. The transformers can make it easier to amplify gate drive power at high frequencies without having to add power supplies at every stage.
subcooledheatpump Thanks for adding the comment. I do understand what you explain. What I have a problem with is how the toroid is actually wound. I am assuming that one lead of each twisted wire is all connected together to form the primary. I really am not sure since I know you can wind several ways, however, only one or two ways will actually work. I have investigated the properties of determining the proper coil use and understand that. When I search for info on how actually a full bridge GDT is wired, I get limited detail with most info on half bridge GDT. I have a question going on another guys youtube channel about this very issue, on his Gate Drive Transformer explanation tutorial which does not show the actual winding in depth. Thank you again and I hope you can share more about this lacking area.
I have seen this; use an Ethernet cable. Wind the cable around the toroid core, use one wire for the primary, and the rest as secondaries. But basically the concept is the same. Wrap around one wire around the core, this is your primary. Any others you wind will be secondaries it's up to you to determine the polarity
subcooledheatpump I think you are saying use one twisted pair of the cable as it's own seperate primary? That would mean you would have to have 2 more twisted wires wound on that toroid and then use those 2 windings as the gate drives. After winding you would have a toatal of 4 twisted pair secondaries, basically 2 twisted pairs on each end of the completed winding with the primary somewhere in there with them. Makes sense but when I look at the schematic on that site, his primary and secondaries show complete isolation. Meaning that all 4 secondaries are isolated. Makes me think that he is using some other way to wind it. The whole study of using a toroid as an isolated GDT for half bridge is fairly easy to make. You can find sites that explain that and even how to get the best results. I would ask you to please clarify how you think those full bridge toroids are actually wound. I appeal to you since you seem to have a good understanding about GDT and are so open to discuss and explain. I really want to know if that full bridge GDT is wound as you put it, or as I have seen elsewhere on other sites in half bridge mode, using twisted pairs but combining one wire in each of those twisted pairs and making those the primary. I know it can be done a few ways, it is the concept of best, "low leakage" and magnetic flux design I am trying to figure out for the full bridge GDT. Thanks Sub.
![[LIVE] : ONE ลุมพินี 65 | คู่เอก "จ้าวเสือใหญ่ vs ผึ้งหลวง"](http://i.ytimg.com/vi/TDpOh5eEJKw/mqdefault.jpg)
Yes you can use gate drive transformers. I would advise to check the output of the transformers with an oscilloscope first before you apply power to the IGBTs. Check for ringing, proper voltage, proper rise times, and proper polarity with the transformers connected to the IGBT gates first before you apply power to the DC Bus. This will save your IGBTs from damage, or possibly from destruction. Thanks for watching
good videos thanks,hope expect more videos soon
Good explanation
Technically speaking you can get away with using 3 isolated supplies. One for each of the upper IGBTs and One for both lower IGBTs. Notice on a full bridge diagram, where the emitters are located on the upper IGBTs; they are in common with the output phase. You can't use one supply for both uppers because you would be shorting out both phases with each other. The lower IGBT have their emitters in common with the DC Bus negative, so you can use one supply. Thanks for watching
Thanks for clarifying that for me.
Yup, I'll first "Rectify + Filter" the Output of Transformer & Then Use it as DC Source (20V) Thanx man...!
Thank you, that was helpful.
very good project
so would the igbt modules work good for a DC to ac inverter. h bridge. and for the DC to dc boost . I'm kinda new at all this. but need to build a DC to ac inverter. to run my home. it just seem a few igbt modules would be simpler than a few dozen MOSFETs. I would like to make it where I can expand the potential of it. like 3kw. then double that 2-3x later as I need more power. I guess what I'm wondering is do they switch fast enough to be used in a spwm inverter. I think its a 23-24khz carrier and 60 hz sine wave. thanks and I enjoy your videos
Hey,
For reverse input voltage protection,I decide to employ diodes in series with each IGBT,will an inductive load be able to discharge when the switches are off since the free wheeling diode will not be conducting?
Nice
Hi mate, love your stuff... I've learnt heaps from your easy-speak approach. I do have a question for you though and I'd love to see how you would explain this. I don't think you've covered this subject in any great detail as yet (correct me if I'm wrong). Could you please cover the effect of gate resistance on the dead time in inverters. ie: The delay caused by gate resistors and the dead time required in real circuits such as inverters. Ta mate.
Thank you for posting this nice tutorial. One question : what are the matching coils and what do they do? Can you kindly explain?
what value gate resistors are you using also looks like you have a resistor across gate to emitter on the igbt whats the purpose (pull down ?) and what value .. cheers
This is very interesting.
thanks bud!
good
Is it possible to have a single power supply if you put resistors on the positive and negative of each gate driver?
Great video, very informative! Can I ask what type and size of capacitors you used for this circuit. Both for the capacitor bank on the igbt and the caps used with the 555 timer to get a ac signal. I'm a little confused about the sizing of them. Thanks again for the great video.
V
Svbbybg hygghhhbbbgbb Cxvc jb hh bbbbbbbhbb b bb.
Hello!! Thanks for all you knowledge you post here. I want to know, if I am going to use an IGBT (CM400HA-24A) as a dc motor speed control from 30 A123 20ah cells must I use isolated dc for the gate driver??
Technically you don't have to if you are only driving an IGBT chopper and not a bridge circuit. But I would anyway
CM400DY-24NF CM400DY-50H CM400DY-66H CM400HA-12H
CM400HA-24A CM400HA-24E CM400HA-24H CM400HA-28H
CM400HA-34H CM400HU-24F CM400HU-24H CM400YE2N-12F
need any other igbt module, please kind contact us.
best regards,
sales@asiantx.com
Okay, thanks for the circuit, but how to build 3 phase fullbridge.
you can just put an extra step in order to get 3 phases, or not
Great tutorial. What is the silver foil doing around those power supply wires? Many thanks for sharing.
interferance maybe, either absorbing it or keeping the + and - wires together to cancel their interferance
So if u wanted to create a induction heater capable of melting metal (i wanna get away from propane). Would you say a full bridge inverter to drive the inductor is better than say a half bridge version (not really understanding the differcen). Been trying to read up on the subject.. but i make knifes.. not induction heaters. ;D
A half bridge circuit will only give half the output voltage. A full bridge circuit will give full output voltage. The output of a full bridge circuit will also be smoother, the currents and voltages will be more steady
Could you share your schematics. I would like to apply it for high power application. What frequency do you use and what is the dead time of the square wave generator? Thanks
Thumbs up :)
Alex
Thanks a lot Man this is awesome! You just cleared my head of some nagging questions I've had for some time now.This might be a dumb question, but are those 555 timers on the power supply?
thank mate to share this video , but I have questions
1.what is the resonant frequency of the tank circuit
2.do you use any feedback control system in it?
He use peobably oscillator who drive igbt. He can use oscillator who have the same freque'cy to tuz LC circuit
Hi! Could you give some information about gate-emiter resistors? I haven't catch...
Hey, One more thing, Will it cause "Short-Circuit" if I use Same Driver Power Supply for "Upper Two IGBT" & Other isolated Second Driver Power Supply for "Lower Two IGBTs" ??
2 Isolated Supply is Enough ??
So mosfet can also used instead igbt
Hey Buddy..! May I use 4 different Transformers also instead of Battery for Driver Supply..( 12-0-12 Volt, 1 Ampere ) ? Anyways this Video is Superb .. Keep it Up.
if you could give me an idea frequency to switch my igbt (1khz - 18khz) and some of the benefits of higher vs lower frequency. I figured out most of the small things I'm trying to figure out how much heat this thing should produce. hearing the switching frequency doesn't bother me I have a currently set at 18 kilohertz but gets warm rather fast. I have a heat sink measurements 2 inches tall 7 inches wide 10 inches long, also I tried emailing you but I'm not sure if you got it or if I'm in your junk bin.
Try mailing me again, but put my youtube username in the subject line. That should help it get to my inbox. You've already figured out the most important part about switching frequency. Higher switching frequencies produce a smoother current output, but they also dramatically increase the switching losses. You can lower the switching frequency to reduce IGBT heating, however if the output current becomes too rough, you may need to add in passives (inductor, capacitor, etc....) to smooth the output current
one more thing the square wave output from the full-bridge has a frequency equal to the frequency of the PWM signal from 555 Timer.
how do we convert this high-frequency signal into a 60Hz signal???
hello can you help me with a project,I have a few igbt from an ups and want to build a power supply, inverter and a charger 192v dc.
could this run at 60 hz. and 120v for a DC to AC power inverter. or as a DC to DC boost stage for a inverter. I am needing to build a inverter to power my off grid home. I am self taught. and learning all I can to build my inverter. for like 24vdc to 120vac. thanks for the videos. they have really helped. but I would like to know more about the brain of the system. what powers the FETs or igbt. I have thought about just using a ic555. but I don't think it will do what I need. I need to regulate output and compensate for battery voltage drop. so its the boost stage I'm in sure about. if you could give any help or advice. I will be grateful and thanks again!
did you succeed ?
Yes you can run something like this at 60 Hz, but that would output a square wave and that isn't what you want. You want something that is running PWM that simulates 60 Hz. The easiest way really, is to get yourself a big VFD and an output filter. Set the output frequency to 60 Hz. Then make a boost converter to feed the VFD DC bus input. Of Course, You can make your own inverter that will work as you want. Several micro controller chips are available that are pre programmed to do the job you have in mind. MC3PHAC is a motor controller chip that can send drive signals to gate drivers to switch IGBTs in the correct pattern, you just need to provide the capacitors, oscillator and resistors. Then it would be the same, just feed the DC bus with your boost converter.
+subcooledheatpump thanks for your reply. I still haven't got to far on my project. but I found a inverter driver board with get drivers. looks like it will work. and I was thinking of using a arduino for the DC boost stage. and was thinking of using a center tap push pull transformer. but to put out 6-8kw. it would be huge. and limited to low frequency. I was also of building a core from iron powder and epoxy. basically use a laminated iron core transformer as a model. or go with a toroid of a decent size. maybe 10-12" and see if it would run at 1khz or so. I would like for it to be oversized. so I can expand the inverters output later. to around 10kw or better. as my battery bank grows. I would love your advice. you seem to really know your stuff! and your videos are a very useful learning tool and I have watched a few multiple times. I have learned lots from them. thanks again
i want to ask sir about resistor value between driver and gate, if the input voltage 240v DC?
Hi bro i want to use semikron igbt skm200gb063d for induction furnace but i don't have SOA graph of that igbt i find in datasheet but that is not there how i can find max power of that igbt can you help pls....
Thank you for your speedy response!! is there a way I could get you to help me with some details?? I would pay you for your time and experience. If not its OK
Sure I can try to help. Send me an email. dhc@dhcservice.com Thanks
hi dear i want build a single phase motor controller i made 2 square wave with SG3525AN ic that i can change the frequency of this two wave 9hz up to 300hz if i want drive 4 igbt with this square wave i have to use the hcpl ic? i can not understand what how can we connect the negative of 15 volt source for vcc of hcpl ic to emitter of igbt because we have the flow of current to load i mean this current dosent harmful for hcpl ic.thank you so much
in an 15n120 igbt +ve gives to collector how to control this
I don't fully understand this about the fully isolated power supply, and isolated grounds, for instance if you have build a linear power supply, does that mean that you have to use 4 transformers, 4 regulators one for each IGBT driver?, or you can use 1 tranformer, 4 regulators? Why must Vcc for the octocoupler be even isolated, isnt' that just power supply for the IGBT. I'm confused. Thanks in advance.
You need isolated flating poWet supply cauz if the power supply is the same for these 4 gates, you make short circuit at the h bridge.
hai !!! nice work ..one question though !! you speak about four ISOLATED power supplies..but on your diagram they all seems to connected by ground bus..please clarify on this..thanks in advance
They are isolated in the diagram. Look closely. What you're seeing aren't the power supplies
Hey, very nice explain, thz gate driver using and why interverting. But, i f i want ro send just cycle lf swiching ( i want these 4 igbt switch off, how i can get that? Is needed to send all at gnd ?
To switch an IGBT off with a gate driver, all that's needed is removal of the input signal to the gate driver. Without a gate driver, simply short the emitter and gate terminals
@@subcooledheatpumpmmh, i think the gate works like capacitor parasite thx for answer
What is the value of the capacitor?
Is it not possible to make a full bridge Inverter with only one power supply for the gate drivers?
Some gate drivers can drive both high and low side and provide a form of isolation that is built in. You can use just one power supply for the low side IGBTs, but the high sides are where the isolation is required.
can we derive four IGBTS with two gate drivers as you know
two IGBTs operate simultaneously in H bridge
plz help me i am in trouble
4 isolated power supplys :/
Just use GDT ?
you are using gate resistor at the output of driver ic.what is the value of second resistor that you have solded on IGBT module.how you have selected that resistor.i assume you have connected that resistor between gate and drain of IGBT.this resistor has not been shown in circuit diagram.is this resistor is to charge and discharge gate capacitance of IGBT module.please correct me.
The place I bought those IGBTs from sometimes would have a 1K resistor soldered between the gate and emitter, this keeps the gate from being damaged by static buildup during shipping. It's also sometimes helpful if you aren't using negative gate drive/bias, just to drain the gate and be sure it won't turn on without a signal from the gate driver
is there any way to calculate the value of resistor connected between gate and emitter by using data sheet.i have build the circuit of induction heater but one of my module is heating while second is normal.is this due to square signal of IGBT gate driver.can I trigger gate using sine wave.if yes then how to implement dead time technique. is it possible to reduce heating by reducing deadtime.i mean is heating effect is deadtime dependent.
hello i want to make three phase inverter for speed control of 3 phase 5 hp indution motor so please tell me whats rating of igbts should i use and also other components please..............
Hey I am working on a similar project. Can you give me a few pointers about it's design?
A full bridge "inverter"? Interesting!
One way of classifying power conversion systems is according to whether the input and output are alternating current (AC) or direct current (DC), thus:
DC to DC = DC-to-DC converter, Voltage regulator, Linear regulator
AC to DC = Rectifier, power supply unit (PSU)
DC to AC = Inverter
AC to AC = Transformer, Voltage converter, Voltage regulator, Variable-frequency transformer
Hi
Dear
This is Farid.
kindly make a single phase AC motor speed control upto 100hz.
Thanks
Sir,
How HCPL3120 input operate ? Do you have any video please ?
can we use single igbt from brick package
Yes you can, but if you're going to do that, make sure to short out the gate and emitter terminals on the unused IGBTs
Hi ! why the snubber is connected to the dc bus ?
The IGBT turn off spike also has an effect on the bus voltage. The main DC bus capacitors won't response fast enough to quench the initial spike. So the metal film capacitors help with that.
@@subcooledheatpump So , i must connect snubber to DC bus , and to switch ?
@@deric1 Unless you're having major problems with turn off spikes, it usually isn't necessary to connect on across the switch itself.
nice dude. also check out this crazy chip. STGIPS30C60T-H . it includes drivers and IGBTs all in one, just needs PWM st right from MCU to chip. only thing needed is your shunt current measurements, the MCU and caps.
How can I reach you?
how to.make 20khz osilitor board?
hello sir i need about igbt detail using this project so please provide me for imporve project thank u
hello sir , can u explain me how can u manage dead band of igbts, means how all igbts are off for some fractions??
That usually isn't managed by the IGBTs or the gate drivers, but by the PWM generator which supplies the gate drivers. Many PWM IC's feature built in dead time control, MC3PHAC is one example of a PWM motor controller which features built in dead time control
Thanks for reply,
Suppose I'm giving pulses to igbts using Microcontroller then it's necessary to give pulses with dead band difference?
Means suppose I'm using h bridge mode at that time I'm switching on diagonally igbt pair and then other pair diagonally...At that time between these two switchings I should put dead band?
Buying all those parts should be pretty expensive. Especially those power IGBTs
hello
I wanna ask about how to connect full bridge inverter using 2 devices of DM2G100SH12AL??
tqvm
DM2G100SH12AL
please kind email sales@asiantx.com
what dead time you have set?is there any limit of primary source that you are chopping? plz mention ur email id,so that I can discuss you my issues.
One question, your video didn't answer is how to choose the value of Rg.(gate driver resistor)
The Rg is very application dependent. IGBT manufactures have a list of generally acceptable Rg values for a given IGBT. In general, Rg must be raised if higher voltages are to be used, or if the turn off voltage spike is too high. Rg must be reduced if the thermal dissipation is exceeded, or high switching frequencies are to be used.
As for the isolated power supplies, the high side (positive) IGBT gate drive power supplies MUST be isolated, unless you like explosions. The low side gate drive power supplies don't need to be, but some are anyway to reduce the possibility of noise or interference
subcooledheatpump Thanks mate for answer. That and some more things, I would only like to propose you, to show in one of your videos, there are not many videos out here, how to drive this IGBTs blocks, I mean the real "basics", showing also the use in practice, Maybe on some power managment too, like 'phase-shift control of H-bridge'(I picked that up from richie brunnet site*) and others.. I greatly appreciate..
My friend you read datasheet for Optocoupler
www.bmh.nu/pdf/Opto/HCPL3120.pdf
The full schematic diagram
I wouldn't recommend to solder part on IGBT brick. It can low lifetime or degrade IGBT by heat. Also it is unsafe. If spark, IGBT might died. Hundred dollar draining. Gate part are not very strong as you think. Spark can happening anytime.
Good Points. These IGBTs are just surplus so I didn't mind abusing them a little to save some time.
tank coil how to connect
Can please specify the dead time
+Tinsae Berhane
Hi mate, was that question directed to subcooledheatpump or myself>
+paul bergin For all the questions about dead time, Dead time is simply the required time when all IGBTs are switched off. For each IGBT, a value is specified in the datasheet for these characteristics; Turn on Delay, Rise time, Turn off Delay, and Fall time. You will notice that most IGBTs usually turn on faster than they turn off. In Inverter circuits, such as shown in this video, this could cause both high and low side IGBTs to be on at the same time, causing a short circuit known as "shoot-through" (IE. lower would switch on before the upper switches off) Dead Time simply turns off both IGBTs for a very short time (usually around 1 microsecond) to allow each IGBT to fully turn off before the opposing IGBT turns on. Gate resistance has more of an impact on turn on time, as the IGBT's gate is a capacitor, higher resistance leads to a slower turn on. This could help relieve some mismatch in turn on and turn off times. It is possible to use two different gate resistors on the same IGBT, one for turn on and a different one for turn off to match the rise and fall times, this helps reduce or eliminate dead time. The gate resistance does affect turn off, but only so much, as IGBTs like other bipolar transistors feature a parasitic "current tail" or fall time. This current tail will limit the IGBTs turn off speed, and is thus the reason inverter circuits need to use some dead time.
+subcooledheatpump
Thanks heaps buddy, this is what I imagined was the case, the difference between the turn on and off times for my Ixys IGBT's is quite noticeable on the scope and would definitely lead to problems with shoot-through as you up the drive frequency. So in general a higher gate resistance will delay the on saturation time which will alleviate the risk of crossover time, I also understand that too much delay will lessen performance. I have set up a test rig to visualize the actual gap between output pulses on my scope whilst running at 20khz. I will aim for 1us. Ta mate.
+paul bergin sorry for the confusion it is for subcooledheatpump
Thank you very much i have the answer for my question. it shows that i can ask more question. Thank you again!
It is used to shield out interference
What is this unit called for and what does it do? Found 2 of these with hot-swap(?) handle.
3 Pictures here: postimg.org/gallery/2gophpmlk/
Than serial number is for transistor module and that way I found your video.
Not sure what type of unit those are, but the transistor is not an IGBT, it's a darlington pair transistor
Thanks man!
please send me igbt gate drive circuit
I'm not understanding why you say each "gate driver" needs an isolated power supply. Both high-side drivers could share a supply, and both low-side drivers could share another... but why not use real gate drivers with charge pump... no worries at all.
Only the low side gate drivers could share a power supply. The high sides need some form of isolation between them, otherwise the bridge output will be short circuited
Therein lies the reason the full bridge charge pump gate drivers exist... like SM7295, A3921, DRV8704 (dual H-bridge), MC33883... and many more. Even the ooollld faithful IR2110 is great for this... it just takes two of them for a full H-bridge.
@@beforebefore Yeah charge pumps are ok for lower power applications. Personally I never had too much luck with them. Probably did something wrong I'm sure. Most manufactures recommend isolated power supplies, even for the low sides after a certain power level.
You lost me at the ac connected to the signal input I never seen a ac connected to a signal output
I like your gate drive. I would ask if you could shed some light on a FullB induction heater listed on instructables.
www.instructables.com/id/30-kVA-Induction-Heater/
If you notice the isolated GDTr's are wound toroids. I have looked at this design and am a bit condused about the details of the windings. You sound like you could clarify that issue and have most likely used that system before. Can you please share anything? Thanks.
The gate drive transformer he used was toroidal, that just means the windings were wound onto a ring like transformer structure. They are electrically the same as any other transformer. As for how they work, there is one primary coil, driven by an amplifier stage. There are four secondaries. with two of positive polarity, and two negative polarity, that is, with one half switching cycle. On the other half of the switching cycle, they reverse. This is used to drive the opposing IGBTs on and off at the right time to avoid shoot through. Basically the same thing as you see in my video, just done with gate drive transformers. The transformers can make it easier to amplify gate drive power at high frequencies without having to add power supplies at every stage.
subcooledheatpump
Thanks for adding the comment. I do understand what you explain. What I have a problem with is how the toroid is actually wound. I am assuming that one lead of each twisted wire is all connected together to form the primary. I really am not sure since I know you can wind several ways, however, only one or two ways will actually work. I have investigated the properties of determining the proper coil use and understand that. When I search for info on how actually a full bridge GDT is wired, I get limited detail with most info on half bridge GDT. I have a question going on another guys youtube channel about this very issue, on his Gate Drive Transformer explanation tutorial which does not show the actual winding in depth. Thank you again and I hope you can share more about this lacking area.
I have seen this; use an Ethernet cable. Wind the cable around the toroid core, use one wire for the primary, and the rest as secondaries. But basically the concept is the same. Wrap around one wire around the core, this is your primary. Any others you wind will be secondaries it's up to you to determine the polarity
subcooledheatpump
I think you are saying use one twisted pair of the cable as it's own seperate primary? That would mean you would have to have 2 more twisted wires wound on that toroid and then use those 2 windings as the gate drives. After winding you would have a toatal of 4 twisted pair secondaries, basically 2 twisted pairs on each end of the completed winding with the primary somewhere in there with them. Makes sense but when I look at the schematic on that site, his primary and secondaries show complete isolation. Meaning that all 4 secondaries are isolated. Makes me think that he is using some other way to wind it. The whole study of using a toroid as an isolated GDT for half bridge is fairly easy to make. You can find sites that explain that and even how to get the best results.
I would ask you to please clarify how you think those full bridge toroids are actually wound. I appeal to you since you seem to have a good understanding about GDT and are so open to discuss and explain. I really want to know if that full bridge GDT is wound as you put it, or as I have seen elsewhere on other sites in half bridge mode, using twisted pairs but combining one wire in each of those twisted pairs and making those the primary. I know it can be done a few ways, it is the concept of best, "low leakage" and magnetic flux design I am trying to figure out for the full bridge GDT.
Thanks Sub.