Inject tone at (R) FET11 Drain = faint, Gate = 0, Source = 0 at Line OUT
Inject tone at (L) FET11 Drain = med, Gate = faint, Source = med at L/OUT
Next I injected a tone at FWD head (R) to SW4-1 = OK same on (L)= OK
(R) Then from SW4-1 to TR9 Coll. = OK same on (L) = OK
(R) from TR9 Coll. to TR10 Coll. = OK same on (L) = OK
(R) from TR10 Coll. to TR11 Source = OK same on (L) = OK
(R) from TR11 Drain to TR12 Coll. = OK same on (L) = OK
(R) from TR11 Source to TR11 Drain = faint same on (L) = OK
Now Voltage at TR's:
Cct.Diag: TR7 e=0.0 c=1.4 b=0.7
Actual: (R) TR7 e=0.0 c=1.0 b=0.3 Actual: (L) TR7 e=0.0 c=1.0 b=0.3
Cct.Diag: TR8 e=0.8 c=6.3 b=1.4
Actual: (R) TR8 e=0.4 c=6.4 b=1.0 Act: (L) TR8 e=0.4 c=6.5 b=1.0
Cct.Diag: TR9 e=0.5 c=1.9 b=1.1
Actual: (R) TR9 e=0.2 c=1.5 b=0.8 Act: (L) TR9 e=0.2 c=1.5 b=0.8
Cct.Diag: TR10 e=1.2 c=8.3 b=1.9
Actual: (R) TR10 e=0.9 c=8.3 b=1.5 Act: (L) TR10 e=0.9 c=8.2 b=0.0
Cct.Diag: FET11 d=21.2 s=21.2 g=0.0
Actual: (R) FET11 d=20.5 g=0 s=20.5 Act: (L) FET11 d=20.4 g=20.3 s=20.4
Cct.Diag: TR12 e=10.6 c=22.4 b=11.2
Actual: (R) TR12 e=9.5 c=20.5 b=10.0 Act: (L) TR12 e=9.5 c=20.6 b=10.1
I hope these table reproduce OK !
I think the (L) FET is shorted out within or that C38 has failed = which could well be the case, thus reinforcing the need to replace it! (or may be more than just C38? but I note it's not an electrolytic) I also note that the tone results when injected at the FET drain is different from (L)FET to (R)FET see first test.
_______________
Michael
Last edited:
OK, you are being systematic there, which is good. The FETs TR11, do appear quite different and I agree that (L) side is suspicious. Check voltages on the other side of the caps and at TR19 collector, for the remote possibility there are other problems but it does look like TR11(L) is shorted internally. Remove and test in your wonder tester! 😉
Replacements could be difficult because JFETS are all over the shop with their performance between batches and from one device to another. It can be expensive to use them but now we all face the virtual end of leaded (through-hole) JFET transistor manufacturing. The only stocks of affordable parts will soon be old stock with no guarantees of quality or source. Meantime the 2SK68 can be substituted with 2SK117GR according to the manual.
Amazon even, seem to have an offer but I know nothing about the quality, though it shouldn't be problem in that location: Amazon.com: 2 pcs OF 2SK117 2SK117GR N-Channel 50V 14mA the transistor: Home Improvement
Replacements could be difficult because JFETS are all over the shop with their performance between batches and from one device to another. It can be expensive to use them but now we all face the virtual end of leaded (through-hole) JFET transistor manufacturing. The only stocks of affordable parts will soon be old stock with no guarantees of quality or source. Meantime the 2SK68 can be substituted with 2SK117GR according to the manual.
Amazon even, seem to have an offer but I know nothing about the quality, though it shouldn't be problem in that location: Amazon.com: 2 pcs OF 2SK117 2SK117GR N-Channel 50V 14mA the transistor: Home Improvement
Thank you very much for the info, Ian. I tried the Amazon site, but they won't post to UK. However I have found and ordered a Pkt of 6 from Ebay 'yoyo_components ( 7946 )' the following: 2SK117-BL TOSHIBA Low Noise Audio Amplifier N-Channel TO-92 J-FET which should arrive, Post Free(!) on Tuesday for a total £6.95 GBP. I shall test them for 'matching' qualities - and if I find two very similar I may well replace both TR11's. I found many of the MOSFET 2SK117's in the wrong packaging - i.e. not TO-92 and wouldn't fit therefore. Now I need to dismantle the Pre Amp Board from the carcase again (groan) - it's that business of the outer-ring pre-select stop on both double-knob Gain controls (Mic & Line) where I had difficulty re-assembling since the black plastic washer-insert to one of the knobs is getting a bit 'friable'. Extra care needed here. Then I shall remove the offending 2SK68 FET and measure it with my 'magic' DCA55. At least I should get some indication of 'fail'. If not . . .
_______________
Michael
Here's what the DCA55 has to say about the two TR11's:
TR11 (L) Common anode diode network:
Pinout: D1 = 2=Anode 3=Cathode
D2 = 1=Cathode 2=Anode
TR11 (R) Drain and Source not identified. Pin 2=Gate
N-Channel Junction FET
So what conclusion can be drawn from this?
Are there DMM tests that can be applied?
______________
Michael
Last edited:
Your Peak meter can identify a JFET but not tell you more than it works or not. Should I add insult by saying the cheapo Ebay meter has no problem with this? I guess not, as Peak's Youtube video already admits this problem. However, you do have confirmation that the suspect doesn't work - so replace it.
Again, there is no point trying to match anything here. The FETs are simply gating a signal on or off and matching is not the issue. All you need is 1 JFET, 2SK68 or 2SK117, grade GR (2,5-6.5 mA IDss) as specified. Other small signal types may be suitable too but you need the GR grade. BL grade is a higher range of current (6-14 mA IDss) and may not operate correctly. Let's not go down the path of testing JFETs which is a half a books worth of study, experimentation and more equipment than a DVM to get started properly. More to the point, you won't benefit from messing about testing anything for the sake of 1 cheap JFET.
Peak Electronic Design Limited - Atlas DCA - Semiconductor Analyser - Model DCA55
https://www.youtube.com/watch?v=zo_BejqYSF8&feature=youtu.be
Again, there is no point trying to match anything here. The FETs are simply gating a signal on or off and matching is not the issue. All you need is 1 JFET, 2SK68 or 2SK117, grade GR (2,5-6.5 mA IDss) as specified. Other small signal types may be suitable too but you need the GR grade. BL grade is a higher range of current (6-14 mA IDss) and may not operate correctly. Let's not go down the path of testing JFETs which is a half a books worth of study, experimentation and more equipment than a DVM to get started properly. More to the point, you won't benefit from messing about testing anything for the sake of 1 cheap JFET.
Peak Electronic Design Limited - Atlas DCA - Semiconductor Analyser - Model DCA55
https://www.youtube.com/watch?v=zo_BejqYSF8&feature=youtu.be
Surely, your local UK supplier, Dalbani won't have difficulty mailing to you: ELECTRONIC COMPONENTS, SEMICONDUCTORS, PARTS, TRANSISTORS, INVERTERS, TRANSFORMERS | DALBANI
Many thanks for this Ian! I didn't know of Dalbani and have opened the required Account and ordered 2 x SK117GR's because I had removed both for testing and I feel it best to replace both with new SK117GR's. I hope you agree! Obviously I'm having difficulties learning the 'ropes' regarding components and I am truly grateful for your help and guidance in this. I wasn't aware of the Grading and its importance. The data sheets I have for NEC 2SK68 and TOSHIBA 2SK117 make no reference to this Grading, so I can only think it's something you either 'know' or (like me) 'don't know'. The BL grade JFETS are in the post now so it's too late to cancel - but, by comparison to the fiasco surrounding the loss of all those KSC1845's a few weeks ago, this is a small loss. I tried using my HoldPeak HP-9804 DMM to test both SK68's but the info I got from the internet for testing FET seemed to have a bit of mumbo-jumbo in: 'touch a finger to both the source and drain pins while continuing to hold the negative probe to the source and the positive probe to the drain . . .' This is from an article: 'How to Test Field Effect Transistors'. As to the DCA55 I only got it to test transistors out-of-Cct. and did not know it had difficulties with FET's. Whilst awaiting the arrival of the 117GR's I shall take my Acer 9300 Laptop to pieces again and apply a heat gun to the VGA chip solder pins on the MoBo - to see if I can get it going again. I am busy in my retirement!
_______________
Michael
As said, FETS are something of a "pot luck" affair in regard to their performance. Any help from manufacturers in having them graded so that they have a better chance of working correctly without the vagaries of odd test methods, is worth the small extra cost, in my view. Such methods may work well enough for experienced individuals but they are not without difficulties for the inexperienced or untrained DIY. We like to think that there will be instruments that solve all component test issues but unless you have a box full of them, you will always find a need for more. Avoid this expensive trap.
The approach to manufacturer grading varies. You are already aware of GR,BL colour name codes and other letters like Y,O,P,R etc used by the Japanese manufacturers. In the US, part numbers may differ but have the same effect. e.g 2N4391,2,3 are graded versions of transistors from the same process and have IDss ranges varying accordingly. Likewise 2N4091,2,3. In Europe though, BF254 came in A,B,C variants for example. As these parts are no more, the SMD equivalents now continue much of the tradition.
As the grading may be achieved by process control or sorting, suffix letters are simply included in a single datasheet. Look at the note, bottom of P1 here:
2SK117GR ... - Datasheet Search Engine Download
The FET117GR's have arrived and been inserted. Now we have new problems!
I have measured the Voltages as they now are through from TR7(R) to TR12(R) and from TR7(L) through to TR12(L) Tables of these follow:
(R)___________________________(L)____________________________
TR7..E=0.54 C=1.24 B= 0.0.........TR7..E=0.54 C=1.22 B=0.00
TR8..E=1.22 C=6.65 B=0.64........TR8..E=1.22 C=6.75 B=0.64
TR9..E=0.97 C=1.71 B=0.39........TR9..E=0.98 C=1.71 B=0.40
TR10 E=1.70 C=8.54 B=1.07........TR10 E=1.71.C=8.45 B=1.08
FET11 D=19.67*G=0.00 S=18.33* FET11 D=14.65*G=0.00 S=15.75*
TR12 E=10.29 C=14.64*B=9.71... TR12 E=10.31 C=14.25*B=9.74
the * at the FET's and TR12 Collectors indicates the voltage at these points disappeared very quickly.
The VU meters no longer indicate any movement whereas they used to hit against the stop on injection of tone.
______________
Michael
There has been a change in the Voltage readings following a restart. These changes affect the voltages given for both (L) and (R) FET's and TR12's:
(R) FET D=20.5 G=0.00 S=20.5 (L) FET D=20.5 G=0.00 S=20.5
(R) TR12 E=9.5 C=20.6 B=10.1 (L) TR12 E=09.5 C=20.6 B=10.1
The Voltages at the FET's and TR12's have stabilized and read as shewn consistently.
The VU meters give full deflection when a tone is injected at either FET Drain.
__________________
Michael
(R) FET D=20.5 G=0.00 S=20.5 (L) FET D=20.5 G=0.00 S=20.5
(R) TR12 E=9.5 C=20.6 B=10.1 (L) TR12 E=09.5 C=20.6 B=10.1
The Voltages at the FET's and TR12's have stabilized and read as shewn consistently.
The VU meters give full deflection when a tone is injected at either FET Drain.
__________________
Michael
Hi again. I can't say what your power hiccup was or whether it may recur, though perhaps being time related, it may well be a power supply capacitor or switch issue. The preamp section looks to be fine for the moment, from your last voltage measurements but as I continue to remind you, REPLACE THE ELECTROLYTICS!
So far, only 1 JFET has been found faulty. However, you must replace the caps ASAP to make sure the hours and cash already spent don't go back up in smoke when the breakdown of any dried out ones wreaks havoc on the other components and brings more than just one or two to a smoky end. That will be much more difficult to sort. Hopefully, you see the point of a handful of replacements being worthwhile insurance against this much worse kind of grief - before being forced to learn the hard way, that is. 😉
Yes indeed! And thanks Ian for reminding me and, for your usual patient guidance. I shall be making an order for replacement caps today, bearing in mind your earlier Post relating to Cap Values and the consequent cost advantage by keeping the Voltage of the caps much the same - where possible. Time is not of essence on my work bench so I forge forward but slowly one step at a time. And this, the Capacitor Step, is now the next step! More fingers X'd!
_________________________
Michael
I am now making a shopping list straight from the Cct.Diag and have a list of 60 caps. I am also investigating Capacitor 'types' and there are several 'Styrol' in the Pre-Amp PCB which seem to have been treated with velvet gloves when soldered in. Two (C29 & C30) were even glued down to the PCB. Should these Styrol caps be replaced? There's a Post on 'AudioKarma" relating to a replacement available from Mouser and made by 'Wima' who maintain "They do not suffer from sensitivity to soldering iron heat the way Styrol caps do". Otherwise I will stick to the Pre/Amp PCB and not touch the Styrol or the Sys.Con.PC Board (yet)
__________________________
Michael
__________________________
Michael
Mouser insist on using Fed-Ex to post your parcel. This doubles the cost of the parts ordered. Still, at least (one hopes) one gets the parts of choice in reasonably fast dispatch!
______________
Michael
______________
Michael
I have found that some of these international couriers ADD ON a handling fee for "importing" even when there is no VAT nor import tax due.
I have had this twice, once via Mouser and once via Digikey.
That extra fee was quite high , ~£11 and ~£23
And that was after Mouser/Digikey had adjusted their prices for sales in the UK and added on their postage costs/fee.
I have had this twice, once via Mouser and once via Digikey.
That extra fee was quite high , ~£11 and ~£23
And that was after Mouser/Digikey had adjusted their prices for sales in the UK and added on their postage costs/fee.
Messrs. 'Wholesale Electronics Inc.' was even higher when I was sourcing some transistors for the GX-635D project - so I didn't even entertain them. Something like $45. or so. But my memory could possibly be at fault.
____________
Michael
Yes, Andrew. I am tempted to replicate the order with Maplin - just to see a/: if they have all the Electrolytics needed b/: How their prices compare. For I know they just use GPO. Fed-Ex in this instance cost me £12.
For electrolytics, Cricklewood electronics seem to be reasonable as are CPC, for retail supplies. I also find it expensive to use Mouser or or Digi-key for electrolytics. Other, more pricey distributors like RS and Farnell can actually be competitive on these items:
Cricklewood Electronics
CPC | CPC - Over 100, 000 products from one of the worlds leading distributors of electronic and related products.
We have already discussed the capacitor types and what to replace earlier, in posts 9-33. So no, you don't need to replace film capacitors like polystyrene or polyesters as they seldom fail in low voltage circuits. Just focus on Electrolytics. If you wish, you can replace the tantalum types in the control circuits with low leakage type electrolytics too but I think it will be less confusing to fit direct replacement types using decent Kemet tantalums on a direct replacement basis. These are not cheap for their sizes, though.
Beware with fitting tantalums and be absolutely sure of correct polarity or you may get a face full of molten tantalum. The markings are minute and often unclear but they can explode violently if mistreated.
Cricklewood Electronics
CPC | CPC - Over 100, 000 products from one of the worlds leading distributors of electronic and related products.
We have already discussed the capacitor types and what to replace earlier, in posts 9-33. So no, you don't need to replace film capacitors like polystyrene or polyesters as they seldom fail in low voltage circuits. Just focus on Electrolytics. If you wish, you can replace the tantalum types in the control circuits with low leakage type electrolytics too but I think it will be less confusing to fit direct replacement types using decent Kemet tantalums on a direct replacement basis. These are not cheap for their sizes, though.
Beware with fitting tantalums and be absolutely sure of correct polarity or you may get a face full of molten tantalum. The markings are minute and often unclear but they can explode violently if mistreated.
Thanks once more, Ian, for your comment. I am definitely (and only) changing the Electrolytics in the Pre/Amp board and for this I have, at vast expense (Fed-Ex freight=£12) purchased from Mouser about £7 worth of selected Electrolytics. My only question relates to the 15 Styrol capacitors. (see Post 93) and I am asking for advice on these. Anyone?
______________________
Michael
- Status
- Not open for further replies.