I have this audio research D130 amplifier, it has some pretty significant damage to one channel. I am really wanting to get this thing up and going, it’s just been kind of sitting on a shelf for a while so I figured let’s take a look. Is there any way of identifying the fats that they use at the input of each channel?
A lot of the components have soot on them, they saw some fire. I know that I can get everything else, I’m going to be replacing the bad outputs with MJL1302/3281. In the service manual it just says “FET color/color/color”.
Is there any possible way of figuring out what these are? Or possibly people that have deciphered this before in the past? I can remove the devices from the good channel to at least give me some specs using a Peak DCA75.
Any help is greatly appreciated
Dan
A lot of the components have soot on them, they saw some fire. I know that I can get everything else, I’m going to be replacing the bad outputs with MJL1302/3281. In the service manual it just says “FET color/color/color”.
Is there any possible way of figuring out what these are? Or possibly people that have deciphered this before in the past? I can remove the devices from the good channel to at least give me some specs using a Peak DCA75.
Any help is greatly appreciated
Dan
If the N-channel and P-channel devices are conplementary: some sort of N-channel FET with a maximum VDS well above 30 V, say 40 V or more, that draws 1.7 mA when VGS = -0.85 V and VDS = 30 V and its P-channel complement.
Thank you for the advice. It looks like they used five different fets, obviously some J and some N. I’m not knowledgeable enough in fits to know exactly how to figure out what you said. I know the VDS, for a get with a VDS of 40v how would the 2SK170 and 2SJ74 work? I have plenty, but unfortunately, I only have them in BL. I do have some others like. j111, 112, 113, 2SK209 (smt) and a couple others, but not their complements I’m sure.
Dan
The fets may still be available from ARC. They were sorted and graded for the circuit.
That would be the best option.
That would be the best option.
I will reach out to them, but I did read some threads where they were charging pretty crazy money for these small TO92 parts. One guy had to pay $30 for one fet. So if I can figure something out that would be nice.
Dan
I hear ya, and I would definitely agree. I’m kind of just hoping, knowing is not likely, that maybe someone has gone through this before and they could be like “oh yeah grey and yellow is an idss of whatever” I can pull the parts from the good channel and measure them with a curve tracer and I’ve measured odds of fets many times. I’ll give them a call next week and see if they would be willing to give me numbers four what each color is specd for. If anyone has any hint as to what these might possibly be, please let me know. I haven’t tried it yet, and it’s not likely, but I’m gonna try and remove the paint and see if there’s any identifiers on the case.
Dan
Some former (and maybe current) ARC employees lurk here.
Maybe one can give you some hints, or even has some parts.
Maybe one can give you some hints, or even has some parts.
Q16 is the utmost remarkable one in this topo...
Running at approx 3Vdc.
Most curious.
The shared nodes of Q18-19 & Q20-21 are highly suspicious, not to say undefined. A flaw?
Running at approx 3Vdc.
Most curious.
The shared nodes of Q18-19 & Q20-21 are highly suspicious, not to say undefined. A flaw?
Last edited:
It looks like some unusual type of SOAR protection, but only for the driver stage, not for the end stage(s).
Because of the voltage, i'd suspect the input JFETs and cascodes are NEC 2SK163 and 2SJ44 parts. These are nice parts, can be better complementary match than 2SK170/2SJ74. The NEC parts have less transconductance but also less capacitance. I don't need to mention how rare they are, right?
If that is the case, I don’t think I will have much luck finding originals. Everywhere I looked though it said that the 2SK170 or 2SK117 and 2SJ74 are excellent replacements for the 163/44. Are you thinking they wouldn’t make appropriate replacements? I am curious how SMT would fare here considering there are so many options vs through hole.
Dan
The only reason these JETs might fail is if the high voltage selection was not good enough. So, either you don't need to replace the JFETs or you need to ~redesign with better (high voltage) parts. You could maybe use MOSFETs or BJTs for the cascode parts. Any amplifier that fails is a bad design and needs modifying.
even many years ago, in the "good ole days" when the good JFETs were plentiful (and relatively cheap), there were very few that would be considered high voltage. As steveu suggested above, if you just have to do something, use a MOSFET or BJT for the cascode device. "i" would only use a JFET on top as a cascode for preamp level voltages (i.e. less than 40V per rail).
at this point in time, whether one JFET is a good replacement for another is almost a useless question because the choices are so limited now. it used to be that as a designer, you could choose for transconductance, capacitance, drain saturation current, package (i.e power dissipation), etc. as your application needed (rayma above very indirectly hints for example that that good parts for a power amp diff stage might not necessarily be the best part for a MC input stage). sadly, this is no longer the case (unless you are jaco, who probably has stock of all parts that ever were ... 🙂), we are pretty much stuck with 2SK170-like parts for better or worse ...
at this point in time, whether one JFET is a good replacement for another is almost a useless question because the choices are so limited now. it used to be that as a designer, you could choose for transconductance, capacitance, drain saturation current, package (i.e power dissipation), etc. as your application needed (rayma above very indirectly hints for example that that good parts for a power amp diff stage might not necessarily be the best part for a MC input stage). sadly, this is no longer the case (unless you are jaco, who probably has stock of all parts that ever were ... 🙂), we are pretty much stuck with 2SK170-like parts for better or worse ...
Last edited:
If you look at the input circuit on the SD-135, it utilizes 2SJ103 / 2SK246. Maybe so here also. "Why change a winning team". The cascode transistors are sourcefollowers and should not be critical. Maybe the same type, differing only in group.
R
R
I think the point is that was a "losing team" and therefore needs replacing. Perhaps it needs ESD protection? IPS failure is very rare and signals something is very wrong.