Mouser Cart for ISP7
Here's the cart for Marks' ISP7 at Mouser.
Remember this is for one board only and you need to check everything against the excel sheet, this will save you a bit of time, but is not intended to think for you!
Click to buy and then add whatever you want, do not edit until then.
Mouser Cart for ISP7
Here's the cart for Marks' ISP7 at Mouser.
Remember this is for one board only and you need to check everything against the excel sheet, this will save you a bit of time, but is not intended to think for you!
Click to buy and then add whatever you want, do not edit until then.
Mouser Cart for ISP7
Thank you for your diligent efforts, @thompsontechs !
I recommend that IPS6 builders purchase AT LEAST 50 pieces of J113 to increase the odds of getting two pairs that match well. I bought 100 pieces myself, see post #3377 of this thread.
The mouser cart above, only includes two pieces of J113
_
Okay, here is an updated cart with the details on location and description as well...
Mouser IPS6
Mouser IPS7
Last edited:
FWIW worth, I built Mark's test jig and tested just the 20 J113s I have available. I found 3 pairs in 20 that were ±5mv or less.
@Mark Johnson: Thanks for describing matching process. It seems to be fairly easy at the price of the J113s to build with closely matched pairs. Now that I have matched pairs it's on to the ISP#6 boards Mark was kind enough to send me.
I've never tried any other JFETs in the IPS6 boards, only the J113s. So I don't know what actually happens in real life, if you substitute 2SK170s.
However I do know that 2SK170s have tremendously higher "gm" (transconductance; gm = dIds / dVgs) than most other JFETs including the J113. So if you put them into the IPS6 amplifier, you're certainly going to get a MUCH higher open loop unity gain frequency "omega_u". See equation (15) of James Solomon's tutorial paper (link) which diyAudio member bonsai graciously hosts at his hifisonix website.
"omega_u" is a relatively advanced concept and I didn't really understand it myself, until my second year of grad school, working towards a master's degree in EE. So if Solomon's paper doesn't click with you right away, don't feel terribly sad. You've moved beyond the academic groove of EE graduate studies; you left the university years and years ago, and it's all rather rusty.
The basic problem is: IPS6 is designed for the gm and the "omega_u" that you get with J113 transistors. Change the transistor, increase its gm, you increase "omega_u" and the amplifier becomes less stable. It could even oscillate! I suspect it does oscillate.
If you've got a dual variable power supply that will go up to ±25 volts, and a pulse generator with a rise time faster than 10 or 15 nanoseconds, and an oscilloscope with a bandwidth of 100 MHz or more, then you can test out the modified circuit yourself, and see how stable or unstable an IPS6 is, with different JFETs. Replace the 100pF compensation capacitor with a socket that lets you plug in lots of different comp caps with different values, and monkey around blindly until you find a capacitor value that gives circuit stability you find acceptable. Good old trial and error! Whey you find something you like: Done!
In my opinion, it's actually a lot quicker to buy 100 J113s, perform the matching exercise, and use those instead. Then you know beforehand, your IPS6 will be stable. The "omega_u" is as-designed, and so is the stability margin-of-safety. 100pF is the right value for the J113 circuit.
Everybody who has built and listened to the IPS6 boards, so far, has used the J113. All the glowing reviews and listener evaluations, came from J113s.
On the other hand, you could decide to design a brand new "IPS6.1" circuit, and you could decide that this new circuit will ONLY use 2SK170 devices. Perform all of the design calculations and circuit topology checks, under these new constraints. When you're all done, your final resulting circuit will either be compatible with the existing IPS6.0 PCB (50% probability) or it won't (50% probability). If it's not, work with one of the PCB-layout-for-hire experts here on diyAudio and get a new PCB design created. Then order some boards, solder them up, test them out, and listen to them in your M2x. It will be tremendous fun!
However I do know that 2SK170s have tremendously higher "gm" (transconductance; gm = dIds / dVgs) than most other JFETs including the J113. So if you put them into the IPS6 amplifier, you're certainly going to get a MUCH higher open loop unity gain frequency "omega_u". See equation (15) of James Solomon's tutorial paper (link) which diyAudio member bonsai graciously hosts at his hifisonix website.
"omega_u" is a relatively advanced concept and I didn't really understand it myself, until my second year of grad school, working towards a master's degree in EE. So if Solomon's paper doesn't click with you right away, don't feel terribly sad. You've moved beyond the academic groove of EE graduate studies; you left the university years and years ago, and it's all rather rusty.
The basic problem is: IPS6 is designed for the gm and the "omega_u" that you get with J113 transistors. Change the transistor, increase its gm, you increase "omega_u" and the amplifier becomes less stable. It could even oscillate! I suspect it does oscillate.
If you've got a dual variable power supply that will go up to ±25 volts, and a pulse generator with a rise time faster than 10 or 15 nanoseconds, and an oscilloscope with a bandwidth of 100 MHz or more, then you can test out the modified circuit yourself, and see how stable or unstable an IPS6 is, with different JFETs. Replace the 100pF compensation capacitor with a socket that lets you plug in lots of different comp caps with different values, and monkey around blindly until you find a capacitor value that gives circuit stability you find acceptable. Good old trial and error! Whey you find something you like: Done!
In my opinion, it's actually a lot quicker to buy 100 J113s, perform the matching exercise, and use those instead. Then you know beforehand, your IPS6 will be stable. The "omega_u" is as-designed, and so is the stability margin-of-safety. 100pF is the right value for the J113 circuit.
Everybody who has built and listened to the IPS6 boards, so far, has used the J113. All the glowing reviews and listener evaluations, came from J113s.
On the other hand, you could decide to design a brand new "IPS6.1" circuit, and you could decide that this new circuit will ONLY use 2SK170 devices. Perform all of the design calculations and circuit topology checks, under these new constraints. When you're all done, your final resulting circuit will either be compatible with the existing IPS6.0 PCB (50% probability) or it won't (50% probability). If it's not, work with one of the PCB-layout-for-hire experts here on diyAudio and get a new PCB design created. Then order some boards, solder them up, test them out, and listen to them in your M2x. It will be tremendous fun!
Ok, thank you for the explanation and the link to the advanced document. I may start out with J113.
I have a Peak DCA 75 Pro small curve tracer. I wonder how good this will be for matching those J113. Probably not as accurate. But I could give it a try.....now I have it.
I have ZTX550 in bulk. I can see that ZTX558 is used in the design. I looked at the data sheets and only real difference I could find is that 550 is a 60V/1A type and 558 is a 400V/0.2A type. Hfe seems similar (100 - 300). So this swap is probably less critical....
I have a Peak DCA 75 Pro small curve tracer. I wonder how good this will be for matching those J113. Probably not as accurate. But I could give it a try.....now I have it.
I have ZTX550 in bulk. I can see that ZTX558 is used in the design. I looked at the data sheets and only real difference I could find is that 550 is a 60V/1A type and 558 is a 400V/0.2A type. Hfe seems similar (100 - 300). So this swap is probably less critical....
Last edited:
I discovered that the output transistors for IPS6 are the same that is used in Wayne 2018 line stage in the "heavy" version. I usually buy a small quantity as the price is so much lower pr. pcs. than buying singles. The problem is now to remember what are "on the shelf" after a number of DIY projects. I guess an excel sheet could help here.....
You might do an analysis of the Risk and the Reward of this parts substitution.
Don, I have 100 on the way, I'm going to jig them up if you qualify through Mark, I would be happy to provide some for you.
J
Yes......if I read "between the lines" the risk is not worth it
OK....ZTX558 is not the most expensive part.
I think the chance that it works with 550 is relative high but if is does not work I will probably regret saving a few cents.....
I have just paid the local PCB manufacture.....and it was quite high compared to the "far east". But then I did some local support and I expect the quality to be high.
J Thompson, I’m willing to contribute the untested parts I purchased towards the cause for a matched pair. I think I only bought 20, but it’s a fair trade (for me anyway) for a matched pair. And then someone else can get a pair from my contribution hopefully. I was going to buy the parts to build the tester this weekend. Happy to trade, though, and Mark already sent me the boards. PM me if interested/willing. Thanks
Last edited:
JT, That is so kind of you. Not only do I get to follow your posts around and "copy" your ideas but now this. I was thinking of doing the same thing. Seems kind of wasteful to have 50 people order 50 J113's and only use a couple.
Thoughts? I'm happy to help or happy to say yes.
Hey, and if Mark says no, I's have a hundred to give away.
He has certainly done the heavy lifting on this one (and others). Kind of makes me feel obligated.The stamps would cost more than the transistors. What does that have to say about tech.
Don
Sure feel free to swap around JFETs all you want, you don't need my permission.
I have a feeling post #3670 means "if you manage to successfully convince MJ to send you free PCBs" ... but I didn't write that post and I might be completely wrong.
Anyway, I have no control over what people do with JFETs they buy or barter, so please do whatever you like. I also have no control over whether people send, or do not send, Gerber files to fabs and buy boards. I also have no control over what people do with the boards they buy, nor what they do with any extra, leftover boards they own. So please do whatever you like, however you like, with whomever you like, and accept my blessings and good wishes.
I have a feeling post #3670 means "if you manage to successfully convince MJ to send you free PCBs" ... but I didn't write that post and I might be completely wrong.
Anyway, I have no control over what people do with JFETs they buy or barter, so please do whatever you like. I also have no control over whether people send, or do not send, Gerber files to fabs and buy boards. I also have no control over what people do with the boards they buy, nor what they do with any extra, leftover boards they own. So please do whatever you like, however you like, with whomever you like, and accept my blessings and good wishes.
joining the club
Few days ago i received my M2x boards. Feels like christmas! This will be my first amp build.
I have found a big used heat sink. I made two halfes out of it, they measures 200mm x 330 mm, each weighs about 2,2 Kg. Hope that will be enough. PSU will be DIY Audio Universal PSU with some big 18000 uF caps and a 500VA toroid.
I will build Ishikawa DB while waiting for parts for the Mountain View DB.
I have soldered the first bits on the mainboard today. Still waiting for mu metal foil to shield the Edcors before soldering them.
Looking forward for the next few weekends of soldering. Thanks to everyone who made this project possible.
Few days ago i received my M2x boards. Feels like christmas! This will be my first amp build.
I have found a big used heat sink. I made two halfes out of it, they measures 200mm x 330 mm, each weighs about 2,2 Kg. Hope that will be enough. PSU will be DIY Audio Universal PSU with some big 18000 uF caps and a 500VA toroid.
I will build Ishikawa DB while waiting for parts for the Mountain View DB.
I have soldered the first bits on the mainboard today. Still waiting for mu metal foil to shield the Edcors before soldering them.
Looking forward for the next few weekends of soldering. Thanks to everyone who made this project possible.
Attachments
J113
I have 100 coming and I don't know that I will get through all 100, so I don't need anymore, but thanks.
I'd like to get enough for my use and then I'll work to get more pairs for others until I go up in smoke from the banality of the task.
Don, I'll get you your pairs after I get going... I have most of the stuff here to get started.
Anyone else, please hit me with a PM.
JT
JT, That is so kind of you. Not only do I get to follow your posts around and "copy" your ideas but now this. I was thinking of doing the same thing. Seems kind of wasteful to have 50 people order 50 J113's and only use a couple.
Thoughts? I'm happy to help or happy to say yes.
Hey, and if Mark says no, I's have a hundred to give away.
The stamps would cost more than the transistors. What does that have to say about tech.
Don
I have 100 coming and I don't know that I will get through all 100, so I don't need anymore, but thanks.
I'd like to get enough for my use and then I'll work to get more pairs for others until I go up in smoke from the banality of the task.
Don, I'll get you your pairs after I get going... I have most of the stuff here to get started.
Anyone else, please hit me with a PM.
JT