As the output relays in my amplifier have started to have issues (contacts degrading) I am going to build the circuit as proposed by Michael Bean as per post 22. I've not really dealt with MOSFETs much so maybe I am missing something, but all data sheets show MOSFETs as having an internal diode/zener already connected from the source to the drain, why would one need to add in a separate external one on top of it?
The FETs I'm thinking of using are these.
The FETs I'm thinking of using are these.
I don't believe they're absolutly necessary, in fact it works just fine without them, I just felt better putting them in. The new MOSFETs that are available are far superior to the IRFZ48V I used in mine, take a look here: http://dkc3.digikey.com/PDF/US2011/1297-1304.pdf
I didn't have to look very long or hard to find some with much lower Rds and higher current and voltage ratings.
Mike
I didn't have to look very long or hard to find some with much lower Rds and higher current and voltage ratings.
Mike
Cool, the nice thing about the FETs is they take up less space then the relays, so I can fit all six channels on one PCB!
Edit - another question. Can you drive 6 channels (12 FETs) from one Photovoltaic Isolator? The gate to source current of 12 FETs should be well below the maximum current output of the isolator, but thinking about it that'd be like connecting the outputs of all 6 channels together, so that'd probably be a bad idea
Edit - another question. Can you drive 6 channels (12 FETs) from one Photovoltaic Isolator? The gate to source current of 12 FETs should be well below the maximum current output of the isolator, but thinking about it that'd be like connecting the outputs of all 6 channels together, so that'd probably be a bad idea
Last edited:
I've had a bit of a browse and the trouble is how they are classified, they are lumped in with solid state relays which makes separating the MOSFET drivers from the complete relays a bit of a challenge. Avago, Panasonic and probably more probably make them. The trouble with the Avago and IR products is they are rather expensive, which is strange because you can buy complete relays for less then £1. You'd figure the driver alone would be the same if not cheaper. The Panasonic one was cheap but out of stock for another month.
The problem is the price, I can buy a replacement relay for 60 pence and replace all 6 for less then £5, currently I'm looking at £5+ per channel. I don't mind spending £1 per mosfet, but I wasn't expecting a dual channel opto thingy to be almost £6 from IR, for that kind of money I can buy a pretty serious relay.
It does however appear that the pricing on the opto whatsits are basically up in the air. The avago ones are considerably cheaper then the IR ones and the Panasonic one cheaper still. I guess it's just a case of weeding out the cheap ones
The problem is the price, I can buy a replacement relay for 60 pence and replace all 6 for less then £5, currently I'm looking at £5+ per channel. I don't mind spending £1 per mosfet, but I wasn't expecting a dual channel opto thingy to be almost £6 from IR, for that kind of money I can buy a pretty serious relay.
It does however appear that the pricing on the opto whatsits are basically up in the air. The avago ones are considerably cheaper then the IR ones and the Panasonic one cheaper still. I guess it's just a case of weeding out the cheap ones
You're probably not going to save a lot of money using these vs mechanical relays (might even cost a little more), but the reliability is far greater, and they're easier to use. I've had mine in an amp I built like 8 or 9 years ago and never had any trouble from them, I can't say that about any of the mechanical buggers I've used.
Mike
Mike
Mooly's irfr3609 has secondary breakdown.
It starts at <7Vds and only allows full SOA for very short pulses, even the 100us pulse cuts off before reaching the full 75Vds.
Is this characteristic common for these low Rds on devices?
or is it unique to IRF?
or is it just this particular FET that shows the breakdown limitation?
It starts at <7Vds and only allows full SOA for very short pulses, even the 100us pulse cuts off before reaching the full 75Vds.
Is this characteristic common for these low Rds on devices?
or is it unique to IRF?
or is it just this particular FET that shows the breakdown limitation?
I have not looked at the spec of the IR device you are referencing. However, Laterals and Verticals are well suited to linear applications, such as the output stages of amplifiers. This is NOT the case with Trench mosfets, which are characterized for switching applications. They do not handle much power in the linear region. MOS Trench Fets can achieve extremely low Rdson and low gate charge. Many vendors also fully characterize their devices for inductive load dump. So, the SOA of these devices is small compared to older technology devices, but they make great switches. Just as an example, there is one vendor offering a 75V device with a 5mO on redirect stance in a TO126 style package.
If one is breaking a 50V supply using these FETs, then the cold SOA determines how quickly the FET must achieve ON to OFF before risking failure of the device. Fig8 shows that 50V breaking a fault current of 1A must be done in less than 1ms when cold. With 10A of fault current to be broken the total switching times becomes very much less than 1ms, but greater than 100us.
If during that short time at very high failure current the device starts to heat up then the temperature de-rated SOA should be used to determine that maximum ON to OFF time.
If the device has high capacitance then fast switching can only be achieved with very high gate current.
This has to do with reliability.
I don't yet know how to switch fast and with high current to achieve that reliability.
By raising the issue I hope I can be pointed in the right direction.
If we think that our supply voltage is 65Vdc and the fault current to be broken is ~15Apk, can the datasheet supply sufficient information to show how quickly the FET must switch off to stay within specification, i.e. reliable?
If during that short time at very high failure current the device starts to heat up then the temperature de-rated SOA should be used to determine that maximum ON to OFF time.
If the device has high capacitance then fast switching can only be achieved with very high gate current.
This has to do with reliability.
I don't yet know how to switch fast and with high current to achieve that reliability.
By raising the issue I hope I can be pointed in the right direction.
If we think that our supply voltage is 65Vdc and the fault current to be broken is ~15Apk, can the datasheet supply sufficient information to show how quickly the FET must switch off to stay within specification, i.e. reliable?
Last edited:
Andrew, I'm not an engineer so I can't provide answers from a position of authority. But from experience, I've tested them to the best of my ability as I said in earlier posts. I too was somewhat sceptical of how well such an arragngement would work, but I could not make them fail, and they had no discernible adverse effect on sound quality.
If it will help ease your mind, heres an excerpt from the PVI5013R data sheet; "The PVI5013R is ideally suited for applications requiring high-current and/or high voltage switching with optical isolation between the low-level driving circuitry and high- energy or high- voltage load circuits. It can be used for directly driving gates of power MOSFETs. The dual- channel configuration allows its outputs to drive independent discrete power MOSFETs, or be connected in parallel or in series to provide higher-current drive for power MOSFETs or higher-voltage drive for IGBTs. PVI5013R employs a fast turn-off circuitry." You can it in its entirety here:
http://www.irf.com/product-info/datasheets/data/pvi5013r.pdf
There are many power MOSFETs available today that I'm sure are rugged enough to be suitable for this application, and you can always parallel devices for increased current capacity.
Mike
If it will help ease your mind, heres an excerpt from the PVI5013R data sheet; "The PVI5013R is ideally suited for applications requiring high-current and/or high voltage switching with optical isolation between the low-level driving circuitry and high- energy or high- voltage load circuits. It can be used for directly driving gates of power MOSFETs. The dual- channel configuration allows its outputs to drive independent discrete power MOSFETs, or be connected in parallel or in series to provide higher-current drive for power MOSFETs or higher-voltage drive for IGBTs. PVI5013R employs a fast turn-off circuitry." You can it in its entirety here:
http://www.irf.com/product-info/datasheets/data/pvi5013r.pdf
There are many power MOSFETs available today that I'm sure are rugged enough to be suitable for this application, and you can always parallel devices for increased current capacity.
Mike
Last edited:
Well here we go a 6 channel solid state relay based DC protection and turn on thump remover.
This is based around the FR3607 MOSFET produced by International Rectifier and the ASSR-V622 MOSFET driver produced by Avago. The delay and detector is based on the design by Rod Elliott. I'd had Rod's design implemented around some Omron sealed relays, but after a year things started to act up.
A few days ago I decided to measure the distortion performance of my blameless 6 channel, just to check that things hadn't gone out of balance, needless to say one channel had gross 3rd order distortion. At the time I didn't bother to investigate why, then a day or two later I noticed that one of the drivers in the active loudspeakers wasn't producing much sound. Turned out it was the contacts in the relays and a day or two later another one stopped working properly too. I had spotted this thread a week or two ago so it couldn't have come at a better time!
The solid state based relay works great and, here's a distortion sweep of the channel that was previously misbehaving.
As you can see, problem solved
This is based around the FR3607 MOSFET produced by International Rectifier and the ASSR-V622 MOSFET driver produced by Avago. The delay and detector is based on the design by Rod Elliott. I'd had Rod's design implemented around some Omron sealed relays, but after a year things started to act up.
A few days ago I decided to measure the distortion performance of my blameless 6 channel, just to check that things hadn't gone out of balance, needless to say one channel had gross 3rd order distortion. At the time I didn't bother to investigate why, then a day or two later I noticed that one of the drivers in the active loudspeakers wasn't producing much sound. Turned out it was the contacts in the relays and a day or two later another one stopped working properly too. I had spotted this thread a week or two ago so it couldn't have come at a better time!
The solid state based relay works great and, here's a distortion sweep of the channel that was previously misbehaving.
As you can see, problem solved
Attachments
That's great, and really interesting to hear you mention the distortion and the actual contact problem effecting driver output audibly.
This has got me wondering you know... months ago I played around with a SPL meter at lowish levels with sine wave testing. Just for curiosity to find nulls etc in the room. Anyway the output from one speaker was "distorted" at certain frequencies. I put the ear plugs in and turned up the level higher and the distortion seemed to reduce and ultimately it dissapeared. On music there was no hint before or after of anything ever being amiss.
Could this be a relay contact issue ?
Oh and well done by the way Excellent job.
This has got me wondering you know... months ago I played around with a SPL meter at lowish levels with sine wave testing. Just for curiosity to find nulls etc in the room. Anyway the output from one speaker was "distorted" at certain frequencies. I put the ear plugs in and turned up the level higher and the distortion seemed to reduce and ultimately it dissapeared. On music there was no hint before or after of anything ever being amiss.
Could this be a relay contact issue ?
Oh and well done by the way
Excellent job.
Hi 5th element,
Wow that was quick, and purdy too! Mine is still on the fugly perf-board setup I kludged together so many years ago, I'm kind of lazy like that, and it still works as good as day one. And it goes to show how great minds think alike, I'm using Rod Elliots DC detect/thump eliminator for the front end of mine too. Good work.
Mike
Wow that was quick, and purdy too! Mine is still on the fugly perf-board setup I kludged together so many years ago, I'm kind of lazy like that, and it still works as good as day one. And it goes to show how great minds think alike, I'm using Rod Elliots DC detect/thump eliminator for the front end of mine too. Good work.
Mike