That's OK. I may just throw these in the cabinet and worry about them later. I have my eye on the Slewmonster at the present so I may just drop this and come back to it when I have more patience.
Blessings, Terry
Blessings, Terry
hey, this is a DIY'ers forum,
there are a lot of other "board artists" here who can do board revisions,
maybe they can help if they have the time... 😀
there are a lot of other "board artists" here who can do board revisions,
maybe they can help if they have the time... 😀
Hi Guys
" Self has a 220uf cap there"
Not in any schematics in his book - all 1mF.
If you use a small value there THD at low frequencies will rise quickly. So, a very sub-audio roll-off point is chosen typically around 1Hz or less - this is flat at 10Hz.
If you make the feedback resistor value lower then C4 can charge more quickly. However, to keep the gain the same R15 has to be made proportionally smaller as well, which then pushes the low-F point and low-F THD higher - then C4 has to be proportionately higher in value.
Cascodes are good and bad, like any other circuit or element. They can increase low-F gain, which should make DC offset control inherently better. On the other hand, they can make high-F instabilities which require compensation of some form or other. Apart from not liking the complexity they add, the high-F issue is why I generally don't use cascodes. In circuits where speed is the goal - like a lot of the CFA amps - you see cascodes everywhere. I've never had an actual problem with a cascode but haven't used them since around '90, when developments in BJTs allowed use of single devices in most places that previously required cascodes just to accommodate voltage.
Both Self and Cordell pointed out pitfalls to using speed-up caps across the driver emitter R, so I leave it out. Having the cross-coupling is usually fast enough - much better than separate Rs to the output node.
As Ranchu pointed out,and as I have said many times, the circuit here is basically the same as any other blameless amp you'll find just with the cascodes added and larger local filters. Self has finally added the latter to some of his newer kits. The bias gen is the same as the other design shown except for how the trimpot is wired. I prefer to wire the trimpot so that if the wiper opens the bias default is to cold.
Unlike Terry's $6 boards, my supplier has a $150 minimum basically, and I did not get boards of this for myself since I have no need for an amp of this power nor do i want to sell boards for this.
Have fun
Kevin O'Connor
" Self has a 220uf cap there"
Not in any schematics in his book - all 1mF.
If you use a small value there THD at low frequencies will rise quickly. So, a very sub-audio roll-off point is chosen typically around 1Hz or less - this is flat at 10Hz.
If you make the feedback resistor value lower then C4 can charge more quickly. However, to keep the gain the same R15 has to be made proportionally smaller as well, which then pushes the low-F point and low-F THD higher - then C4 has to be proportionately higher in value.
Cascodes are good and bad, like any other circuit or element. They can increase low-F gain, which should make DC offset control inherently better. On the other hand, they can make high-F instabilities which require compensation of some form or other. Apart from not liking the complexity they add, the high-F issue is why I generally don't use cascodes. In circuits where speed is the goal - like a lot of the CFA amps - you see cascodes everywhere. I've never had an actual problem with a cascode but haven't used them since around '90, when developments in BJTs allowed use of single devices in most places that previously required cascodes just to accommodate voltage.
Both Self and Cordell pointed out pitfalls to using speed-up caps across the driver emitter R, so I leave it out. Having the cross-coupling is usually fast enough - much better than separate Rs to the output node.
As Ranchu pointed out,and as I have said many times, the circuit here is basically the same as any other blameless amp you'll find just with the cascodes added and larger local filters. Self has finally added the latter to some of his newer kits. The bias gen is the same as the other design shown except for how the trimpot is wired. I prefer to wire the trimpot so that if the wiper opens the bias default is to cold.
Unlike Terry's $6 boards, my supplier has a $150 minimum basically, and I did not get boards of this for myself since I have no need for an amp of this power nor do i want to sell boards for this.
Have fun
Kevin O'Connor
The schematic showing a 220uf cap for C4 is here
http://www.douglas-self.com/ampins/dipa/dpafig33.gif
I'm not saying that he is right. I just know that my offset starts very positive and then settles to a lower negative offset after several seconds with a 40mF cap there. I don't understand how it works, I'm only reporting what I read on my meter.
I'm not saying the circuit doesn't work. It works in spice. What I'm saying is that it doesn't work on the boards built from your Gerbers. I did my best to go over every trace and mark them on the schematic with a highlighter. It looks like they are correct aside for the problem traces at the vbe that we discovered earlier.
I don't need another amp this size either. I built this because after following your progress in this thread, I was anxious to try it and see if it really was better than the MKIII and Leach Superamp, which were the other amps suggested to brlmat. I will never know that if I can't have a working amp.
Even though the boards were $6, I believer they faithfully followed your Gerber files.
I need to go back through my boards now because I tried a few things along the way trying to get rid of the offset. I'd love to get this amp working if for no other reason than for you to be able to fix the Gerbers so others can build a working amp.
Blessings, Terry
Am I getting this right, "Struth(Kevin)" released gerber fab data, of a pcb design that Terry is trying make the first proto work?
Hi Guys
It is surprising that such low values are shown on his site since that info is based on the original Wireless World articles, as are the books. Everywhere else he uses 1mF as a minimum, and demonstrates clearly why high values should be used.
I assume your 40m is a typo as the recommended value is 2m2 to 4m7 (m=milli).
Have fun
Kevin O'Connor
It is surprising that such low values are shown on his site since that info is based on the original Wireless World articles, as are the books. Everywhere else he uses 1mF as a minimum, and demonstrates clearly why high values should be used.
I assume your 40m is a typo as the recommended value is 2m2 to 4m7 (m=milli).
Have fun
Kevin O'Connor
Sorry, yes, I meant 4700uf. I used 2200uf in the third board. I just changed one of them to 330uf and another to 220uf to try out. The offset settles way quicker with those. As you know, I don't have any electronics background. I rely on you folks here to get things working. I have been trying things as guys suggest "fixes". The amp is not working. I don't want to stop until it does. Otherwise all of this time, money and effort was for nothing.
Blessings, Terry
Blessings, Terry
Struth should have an opportunity to assist with debugging since he should now have a set of boards.
Hopefully Kevin will make the effort and populate a board and test it. I have set mine aside for now and worked on other projects. I got tired of replacing parts with no or worse results. Thanks Jason.
It has been quite a while since Kevin has had the boards that Jason sent to him. He was very good at replying to emails for a time but has not returned my last three or four attempts. Maybe he was just unable to get this circuit to work and wants it to just fade into the past. I have quite an investment at this point and would really like to either find that it can work or just can't. I'm posting in hopes that he, (Kevin), gets a notification and peeks in. Maybe my emails are not getting through. I have recently built 6, Slewmaster OPS boards that are EF3 and they work flawlessly so I know EF3 can work. It would sure be good to get to the bottom of these. I hate having to strip boards to salvage parts.
Blessings, Terry
Blessings, Terry
Evening terry.. It's been some time I looked into the forum just been busy with things..
How's the leach Amps doing after all the trouble shooting you had with them..
Regards for now
A..
How's the leach Amps doing after all the trouble shooting you had with them..
Regards for now
A..
Hi Anthony,
The Leach amp are playing fine but it it is likely the super amp will give up its home to a high voltage Slew monster. The Leach served its purpose as a great learning tool but with the beating the home etched boards took, I worry it will not be reliable over the long haul. I have already built a pair of OPS boards to handle the 90v rails. I need the transformer and heatsinks.
I'd still like to get this LTT4 working. I have a little better understanding of how things work together because of the Slewmaster project so maybe I can separate the IPS and vas/ops and test separately. I just received some Symasui IPS boards so I will be working on those for a couple days. I'll keep watching here in hopes St Ruth shows up with some new solutions.
Blessings, Terry
The Leach amp are playing fine but it it is likely the super amp will give up its home to a high voltage Slew monster. The Leach served its purpose as a great learning tool but with the beating the home etched boards took, I worry it will not be reliable over the long haul. I have already built a pair of OPS boards to handle the 90v rails. I need the transformer and heatsinks.
I'd still like to get this LTT4 working. I have a little better understanding of how things work together because of the Slewmaster project so maybe I can separate the IPS and vas/ops and test separately. I just received some Symasui IPS boards so I will be working on those for a couple days. I'll keep watching here in hopes St Ruth shows up with some new solutions.
Blessings, Terry
Hi Terry
If you want to get this board up and running you only need to add base stoppers to the outputs. I used some 2R2 1206 surface mount parts at hand and could be grafted onto the board. If you are buying parts then I would purchase 4R7 1206 parts for better stability margin.
Depending on your choice of driver and pre-driver transistors, you may need to add base stoppers and/or collector decoupling into the drivers. I didn't need to but I haven't done anything other than check that it doesn't oscillate on the workbench, and plays music. Long term testing might paint a different story. Look at how it is done in the SlewMaster schematic.
The output stage here is pretty much identical to the SlewMaster, the differences being the Slew has the extra stability enhancing components, the LTT4 has VI limiting - neither of which should alter the sound if set up properly.
If I were you I would salvage whatever components you can and build up a Wolverine IPS. Wolverine + SlewMaster is a close match to LTT4.
If you want to get this board up and running you only need to add base stoppers to the outputs. I used some 2R2 1206 surface mount parts at hand and could be grafted onto the board. If you are buying parts then I would purchase 4R7 1206 parts for better stability margin.
Depending on your choice of driver and pre-driver transistors, you may need to add base stoppers and/or collector decoupling into the drivers. I didn't need to but I haven't done anything other than check that it doesn't oscillate on the workbench, and plays music. Long term testing might paint a different story. Look at how it is done in the SlewMaster schematic.
The output stage here is pretty much identical to the SlewMaster, the differences being the Slew has the extra stability enhancing components, the LTT4 has VI limiting - neither of which should alter the sound if set up properly.
If I were you I would salvage whatever components you can and build up a Wolverine IPS. Wolverine + SlewMaster is a close match to LTT4.
Hi Christian,
I already have a Wolverine IPS and Slewmaster OPS. Sounds awesome as you might expect. I also have the Spooky/Leach IPS, 4-CFA IPS and am building the Symasui IPS right now. Just waiting on a few caps I didn't have in the bin.
I know the Slewmaster is rock solid, well, except for with that Low TIM hybrid. Valery will be building the Slemaster OPS soon and hopefully resolve those issues.
I'm hoping to "prove" that Kevin was right and this amp can work without the base stoppers. It doesn't act like oscillation as far as I can tell. Somewhere, one or more of the devices is not getting enough current to turn on fully. I don't really want to try and salvage parts. Right now I have 12 pair of MJL21193/94 outputs soldered into these three boards. No easy feat removing those without damage. Besides, I'm not building this because I need another big amp. I'm building it to try and learn. If I end up with a nice amp I can pass on to someone, all the better. So my hopes are that Kevin will get back involved in this an help us get it working.
Blessings, Terry
I already have a Wolverine IPS and Slewmaster OPS. Sounds awesome as you might expect. I also have the Spooky/Leach IPS, 4-CFA IPS and am building the Symasui IPS right now. Just waiting on a few caps I didn't have in the bin.
I know the Slewmaster is rock solid, well, except for with that Low TIM hybrid. Valery will be building the Slemaster OPS soon and hopefully resolve those issues.
I'm hoping to "prove" that Kevin was right and this amp can work without the base stoppers. It doesn't act like oscillation as far as I can tell. Somewhere, one or more of the devices is not getting enough current to turn on fully. I don't really want to try and salvage parts. Right now I have 12 pair of MJL21193/94 outputs soldered into these three boards. No easy feat removing those without damage. Besides, I'm not building this because I need another big amp. I'm building it to try and learn. If I end up with a nice amp I can pass on to someone, all the better. So my hopes are that Kevin will get back involved in this an help us get it working.
Blessings, Terry
Sounds to me as though you might be grappling with a different problem. If you can post you "as-built" schematic with voltages we can try to work through it. I must say though that once we solve this immediate problem, suspect oscillation will rear its ugly head as it did for me.
Knowing what I know now I wouldn't lay out an EF3 without output base stoppers at a minimum. Even if it is stable on the workbench without, I'd be suspicious of it bursting into oscillation under different conditions.
Knowing what I know now I wouldn't lay out an EF3 without output base stoppers at a minimum. Even if it is stable on the workbench without, I'd be suspicious of it bursting into oscillation under different conditions.
Hi Christian,
Yes, you would have to read back through the thread. It is all there. I haven't even bothered with it for quite a while but was reorganizing my shop and came across the boards. I had hoped that Kevin would make some kind of effort to get his design working. Now it's looking like that will not happen so once I finish this Symasui I will probably try to dig into it again. I was comparing it to Self's Blameless and it is basically the same aside for the predrivers and a couple of other additions that Kevin said he put there as protection since the OP wanted to use it for PA work of some sort. Anyway, the Blameless doesn't used base stoppers so it will be interesting to see if this can work without them. We'll see.
Blessings, Terry
Yes, you would have to read back through the thread. It is all there. I haven't even bothered with it for quite a while but was reorganizing my shop and came across the boards. I had hoped that Kevin would make some kind of effort to get his design working. Now it's looking like that will not happen so once I finish this Symasui I will probably try to dig into it again. I was comparing it to Self's Blameless and it is basically the same aside for the predrivers and a couple of other additions that Kevin said he put there as protection since the OP wanted to use it for PA work of some sort. Anyway, the Blameless doesn't used base stoppers so it will be interesting to see if this can work without them. We'll see.
Blessings, Terry
Struth is his Forum Name.
Don't confuse the naming issue by introducing other names.
diyAudio - View Profile: Struth
Don't confuse the naming issue by introducing other names.
diyAudio - View Profile: Struth
- Status
- Not open for further replies.