Hello, I'm now coming back to my next problem circuit. This one is downright embarrassing--a Rangemaster, on an Aion PCB. Should be dead simple. It's a Japanese NOS PNP transistor from Small Bear specifically tested to be suitable in a Rangemaster, and the PCB has a charge pump to take standard center negative power supplies. I tested the transistor before install and it tested fine (don't have the values written down but it was in the Rangemaster zone). Schematic is on the Aion BOM but I've also pasted a link here: (https://www.tdpri.com/attachments/upload_2019-3-31_14-38-32-png.597059/)
I cannot get it to bias properly. I am supposed to get voltages at the test points somewhere from -6.8 to 7-7.1V. Currently, closest I can get is -7.77 volts at the test point, with the trim pot at the opposite end, it goes to -7.88V.
Here is what I have tried so far: I initially used the recommended resistor values suggested by small bear (Rb=47k, Re=4.3k) but that didn't work well. So I tried the values suggested in the Aion BOM, but still no luck. The trimpot that I put in didn't quite go up to 5k so I tried a 1.5k value in R3 to try to get more resistance. Aion had said that I might need to try 47k in R2, but that didn't work either. So TLDR, I have the stock Aion build, but with a 1.5k resistor in R3.
I'll post my PCB shortly below. Any ideas? I feel like this should be the easiest build in the world! :icon_cry:
Here are pictures. Although the transistor orientation looks a little weird when you look at the legs, I verified the C B and E on a component tester and lined them up with the C B and E pads on the PCB. So the orientation should be right.
(https://i.postimg.cc/sQvBCWy6/IMG-0391.jpg) (https://postimg.cc/sQvBCWy6)
(https://i.postimg.cc/LnzJS2Kp/IMG-0393.jpg) (https://postimg.cc/LnzJS2Kp)
Hi. What does it sound like? If it sounds like out should or if it sounds good to you then does it matter?
Quote from: cnspedalbuilder on February 14, 2020, 01:58:05 AM
I cannot get it to bias properly. I am supposed to get voltages at the test points somewhere from -6.8 to 7-7.1V. Currently, closest I can get is -7.77 volts at the test point, with the trim pot at the opposite end, it goes to -7.88V.
Than shouldn't happen..
Try to replace R3
and BIAS trimpot with a resistor of an arbitrary value of 3k3 - 3k9 to see what you get..
Put the stock value on the emitter and bias with the base 68k resistor.
you haven't given us the base and emitter voltage readings yet.
Thanks everyone:
Base Voltage: ~ -136mV
Emitter Voltage: -86mV
Antonis and Mozz, I can try those steps, but is it unusual that it's not able to bias with the values in the build doc?
Had this happen to me with a RM build on a BYOC board. Low tech solution: I installed the battery clip & used a battery with ~8v. Biased pretty close & sounds great. Just a thought,
T
Thanks, ideally I'd like to put this on my pedalboard, so I'm hoping to avoid a battery.
Also, I forgot to mention this earlier but in my initial testing the circuit sounded terrible and actually cut the volume a little. That was before several iterations, but given that the bias is still not right, I'm pretty sure it won't sound good.
Quote from: cnspedalbuilder on February 14, 2020, 03:00:30 PM
Thanks everyone:
Base Voltage: ~ -136mV
Emitter Voltage: -86mV
Antonis and Mozz, I can try those steps, but is it unusual that it's not able to bias with the values in the build doc?
if you sling a 470k resistor to supply, and a 6k8 resistor to ground, the midpoint will sit at about 0V13.
if, however, you were wanting to fit a 470k and a 68k, your
base voltage junction would sit at about 1V14 instead. check very carefully your colour bands.
Duck, I pulled the R1 and checked color codes on R1 and R2 and they check out to 470k and 68k. Continuity also checks out. One odd thing is that the resistances measure differently in the circuit than they do when the components are out.
I'm puzzled. It looks like the power supply is working properly, right? I wonder if the problem has to do with the transistor legs. I hooked them up according to my component tester, but the base leg is on the opposite side from what is indicated on the PCB.
Alternatively, is it just the case that you get these weird voltages sometimes depending on the transistor? Should I just socket a pair of resistor pads and try out different values to see what works? If so, which one is best to try?
Thanks everyone.
added info: re-verified the transistor and CB&E are in right spots according to component tester.
HfE=45
Uf=54mV
Some component testers have difficulties with germanium transistors and do not show the correct results. It is not uncommon for them to transpose the collector and emitter connections.
What is the type number (any markings ) of your transistor and can you see any markings that indicate the manufacturer ?
(http://www.muzique.com/lab/images/tran05.jpg)
what's that you say? a japanese transistor from small bear? in that case, don't mess about. see the above pic, yours WILL match. in the pic you can see a little tab - this called the emitter tab. often missing from TO-1 cases, but replaced with a collector dot - a dot of paint up the side where collector resides. and the base will be at the point of the triangle. this rule holds well.
the plastic pack BC series transistors also follow this pattern [with suffix denoted exceptions], either triangle lead layout, or the flat of the body to the hypotenuse of the triangle.
What?? A component tester can get confused between C, B, & E?!?!? Interesting!
OK the transistor is labeled 2SB77. Duck Arse, it does not have a tab or a dot. This appears to be the datasheet:
http://www.datasheet39.com/PDF/824089/2SB77-datasheet.html (http://www.datasheet39.com/PDF/824089/2SB77-datasheet.html)
It says it has a "collector mark" but I'm not sure what to look for. There is an A (or isocelese triangle, not sure which) inside a circle on one side.
I'm not 100% sure but it appears that the C and E are switched relative to what I see on the datasheet.
addendum: On closer inspection, I think the orientation is correct.
Quote from: cnspedalbuilder on February 16, 2020, 11:34:40 AM
it does not have a tab or a dot. This appears to be the datasheet
It says it has a "collector mark" but I'm not sure what to look for
The mark is on the underside of the rolled rim under the metal case as is seen in ducks picture.
Does it not have that?
One more silly question; What is the actual voltage being produced on pin 5 of the TC1044 ?
What voltage does your meter measure it at ?
Go back to stock values 68k and 3.9k. Set trimmer to zero ohms or jumper it. Pick the base resistor, 68k, raise value to 75k or 82k, 100k, see what that does, lower value, see what that does.
Here is a Japanese transistor viewed from the bottom.
(https://i.postimg.cc/HcdgkdB3/20200216-155843.jpg) (https://postimg.cc/HcdgkdB3)
I see now it's a different package.
No painted dots or indents in the can?
Hmmmmm
What does the seller say?
hFE 45 is low for a Rangemaster. Change the pot from 10k to 20k (22k, 25k), and I bet you'll bias better.
Quote from: mozz on February 16, 2020, 04:06:01 PM
Go back to stock values 68k and 3.9k. Set trimmer to zero ohms or jumper it. Pick the base resistor, raise value, see what that does, lower value, see what that does.
Here is a Japanese transistor viewed from the bottom.
(https://i.postimg.cc/HcdgkdB3/20200216-155843.jpg) (https://postimg.cc/HcdgkdB3)
Mozz, thanks that fits with the datasheet I saw, so it looks like the orientation is right. By Base resistor, do you mean R1 on the schematic (currently 470k)?
Quote from: Slowpoke101 on February 16, 2020, 03:15:32 PM
One more silly question; What is the actual voltage being produced on pin 5 of the TC1044 ?
What voltage does your meter measure it at ?
So before I was testing the circuit with a battery, but this time I checked with my 9V OneSpot. voltage at pin 5=-9.49, Collector voltage=8.99.
Quote from: willienillie on February 16, 2020, 04:27:44 PM
hFE 45 is low for a Rangemaster. Change the pot from 10k to 20k (22k, 25k), and I bet you'll bias better.
Does the Range pot affect the bias?
Quote from: duck_arse on February 15, 2020, 08:33:13 AM
Quote from: cnspedalbuilder on February 14, 2020, 03:00:30 PM
Thanks everyone:
Base Voltage: ~ -136mV
Emitter Voltage: -86mV
Antonis and Mozz, I can try those steps, but is it unusual that it's not able to bias with the values in the build doc?
if you sling a 470k resistor to supply, and a 6k8 resistor to ground, the midpoint will sit at about 0V13. .
I keep thinking about this. It seems like the base voltage should be around -1V, right? Something doesn't add up here.
Quote from: cnspedalbuilder on February 16, 2020, 08:57:13 PM
Quote from: willienillie on February 16, 2020, 04:27:44 PM
hFE 45 is low for a Rangemaster. Change the pot from 10k to 20k (22k, 25k), and I bet you'll bias better.
Does the Range pot affect the bias?
Just to be clear, I'm talking about the volume pot ("Boost" in the schematic you posted). It serves as the collector resistor, and totally relates to bias. I got the idea of the larger value pot from RG, I think it's somewhere in this sideways .pdf:
http://www.geofex.com/article_folders/rangemaster/atboost.pdf (http://www.geofex.com/article_folders/rangemaster/atboost.pdf)
It worked for me when I was building one with a similar transistor. I guess I probably used a 25k pot, since that's more common than 20k or 22k, in the US at least. That rangemaster is long gone, sold to a friend. I built another for myself with a 2SB175, hFE around 70, and low leakage. It biases up nicely with the stock circuit values.
Quote from: cnspedalbuilder on February 16, 2020, 11:34:40 AM
What?? A component tester can get confused between C, B, & E?!?!? Interesting!
OK the transistor is labeled 2SB77. Duck Arse, it does not have a tab or a dot. This appears to be the datasheet:
http://www.datasheet39.com/PDF/824089/2SB77-datasheet.html (http://www.datasheet39.com/PDF/824089/2SB77-datasheet.html)
It says it has a "collector mark" but I'm not sure what to look for. There is an A (or isocelese triangle, not sure which) inside a circle on one side.
I'm not 100% sure but it appears that the C and E are switched relative to what I see on the datasheet.
the few Hitachi ge's I can scope here have B in a circle over the collector, but the one Matshushita here has a circle and something hovering, indicating ..... nothing, really. but go with the image posted by mozz, that is the only triangle [pin layout] you need worry about.
Quote from: duck_arse on February 15, 2020, 08:33:13 AM
Quote from: cnspedalbuilder on February 14, 2020, 03:00:30 PM
Thanks everyone:
Base Voltage: ~ -136mV
Emitter Voltage: -86mV
Antonis and Mozz, I can try those steps, but is it unusual that it's not able to bias with the values in the build doc?
if you sling a 470k resistor to supply, and a 6k8 resistor to ground, the midpoint will sit at about 0V13. if, however, you were wanting to fit a 470k and a 68k, your base voltage junction would sit at about 1V14 instead. check very carefully your colour bands.
the circuit shows 470k and 68k. this
tells the base to be about 1.14 V. if the base
is not at that voltage, one of the resistors is wrong [your measure with a correct oriented trannie would suggest 470k and 6k8], or the transistor is ..... not right in some way. I'm not smart enough to say what the wrong connectioned voltages would be.
Quote from: duck_arse on February 17, 2020, 08:48:48 AM
the circuit shows 470k and 68k. this tells the base to be about 1.14 V. if the base is not at that voltage, one of the resistors is wrong [your measure with a correct oriented trannie would suggest 470k and 6k8], or the transistor is ..... not right in some way. I'm not smart enough to say what the wrong connectioned voltages would be.
Can y'all help me decide the best next step? I have been getting suggestions to tweak resistor or pot values to try to get the biasing in the zone. It sounds like you are saying that is a doomed approach because there is either something wrong with the resistor values at the base or the transistor orientation is off.
Here are the possible steps to try, starting from least invasive to most likely to destroy things given my limited desoldering skills:
1. Replace either R1, R2, or R3 with sockets and experiment with different resistor values until the bias voltage comes out right.
2. Desolder and remove the transistor and check the base voltage with transistor removed.
3. Desolder and remove the 10k boost pot and try higher pot values.
In case it helps with the decision, here is another datapoint--when I measure resistance on the meter, the resistor values do not look right when they are in the PCB, but the color bars on the resistors look like they are the right values and at least R1 measured right when I pulled it out of the circuit. I don't know why the effective resistance looks different in the circuit.
Quote from: cnspedalbuilder on February 17, 2020, 01:10:14 PM
Quote from: duck_arse on February 17, 2020, 08:48:48 AM
the circuit shows 470k and 68k. this tells the base to be about 1.14 V. if the base is not at that voltage, one of the resistors is wrong [your measure with a correct oriented trannie would suggest 470k and 6k8], or the transistor is ..... not right in some way. I'm not smart enough to say what the wrong connectioned voltages would be.
Can y'all help me decide the best next step? I have been getting suggestions to tweak resistor or pot values to try to get the biasing in the zone. It sounds like you are saying that is a doomed approach because there is either something wrong with the resistor values at the base or the transistor orientation is off.
Here are the possible steps to try, starting from least invasive to most likely to destroy things given my limited desoldering skills:
1. Replace either R1, R2, or R3 with sockets and experiment with different resistor values until the bias voltage comes out right.
2. Desolder and remove the transistor and check the base voltage with transistor removed.
3. Desolder and remove the 10k boost pot and try higher pot values.
In case it helps with the decision, here is another datapoint--when I measure resistance on the meter, the resistor values do not look right when they are in the PCB, but the color bars on the resistors look like they are the right values and at least R1 measured right when I pulled it out of the circuit. I don't know why the effective resistance looks different in the circuit.
From Aion:
"Germanium transistor: Any PNP germanium transistor with
a gain (HFe) between 65 and 100 will work here. The exact model is not important (and even varied in the original unit)."
Looks like your issues, as I believe a few have said already, are purely biasing issues. Your transistor gain is too low for this circuit. I would grab a new transistor or tweak the bias resistor values and/or the Boost pot to get things in line.
One last thing to keep in mind: Steve from Small Bear describes those as being unsorted raw stock, and it looks like you got one that was pretty low for a Rangemaster. A good thing to do would be to breadboard the circuit and determine what (if any) combination of bias resistors and boost pot get you the right voltages then put those values in the circuit and box it up.
How did the pedal sound?
QuoteIt's a Japanese NOS PNP transistor from Small Bear specifically tested to be suitable in a Rangemaster
if this is one of the japanese for rangemaster dealios, can you show a photo of the page that came with the transistor please? it should have some values written on it. (when I looked his site last night the included page was showing an
npn on the circuit dia. THAT could confuse a few people.)
as to your possible steps - remove the transistor from the board. measure all the resistor values [POWER OFF, obvs] just to check, again. rebend the transistor legs to the match that mozz image. plug the transistor into your
breadboard with the small bear resistor values listed. measure your e-b-c voltages.
and as Big Monk sez - How did the pedal sound? doesn't matter the numbers if the toans are there good.
Quote from: duck_arse on February 18, 2020, 08:18:32 AM
QuoteIt's a Japanese NOS PNP transistor from Small Bear specifically tested to be suitable in a Rangemaster
if this is one of the japanese for rangemaster dealios, can you show a photo of the page that came with the transistor please? it should have some values written on it. (when I looked his site last night the included page was showing an npn on the circuit dia. THAT could confuse a few people.)
as to your possible steps - remove the transistor from the board. measure all the resistor values [POWER OFF, obvs] just to check, again. rebend the transistor legs to the match that mozz image. plug the transistor into your breadboard with the small bear resistor values listed. measure your e-b-c voltages.
and as Big Monk sez - How did the pedal sound? doesn't matter the numbers if the toans are there good.
I think is it Duck:
http://smallbear-electronics.mybigcommerce.com/2sb77-hitachi/
Looks like one of those unsorted lots Steve has with goodies if you get the right one.
I was thinking it was this one ....
http://smallbear-electronics.mybigcommerce.com/transistor-the-rangemaster-pnp-japanese/
Quote from: duck_arse on February 18, 2020, 09:19:26 AM
I was thinking it was this one ....
http://smallbear-electronics.mybigcommerce.com/transistor-the-rangemaster-pnp-japanese/
Duck is right that is the one. It was came with that Dallas Rangemaster sheet and is listed as: "PNP JAPANESE", Rb=47k, Re=4.3k. The plastic component bag says "79" and then below that "L=257". I have no idea what the writing on the packet means, maybe 79 is the Hfe that Steve measured? The value I got was much lower (45), but maybe that's because I am using a cheap component tester.
but I did initially try the values that Steve suggested on the schematic, and those did not work. With those values, the Rangemaster sounded terrible. There was signal loss, not boost. I haven't tried plugging it in after doing all the changes I've described above, so I can certainly try that. I'll also desolder that transistor :icon_cry: and try it on the breadboard.
Thanks to everyone for their help so far.
Quote from: cnspedalbuilder on February 18, 2020, 11:10:24 AM
Quote from: duck_arse on February 18, 2020, 09:19:26 AM
I was thinking it was this one ....
http://smallbear-electronics.mybigcommerce.com/transistor-the-rangemaster-pnp-japanese/
Duck is right that is the one.
Ok. I read one of your other posts where it seemed like you were referencing another transistor.
Quote from: cnspedalbuilder on February 18, 2020, 11:10:24 AM
It was what came with that Dallas Rangemaster sheet and is listed as: "PNP JAPANESE", Rb=47k, Re=4.3k. The plastic component bag says "79" and then below that "L=257". I have no idea what the writing on the packet means, maybe 79 is the Hfe that Steve measured? The value I got was much lower (45), but maybe that's because I am using a cheap component tester.
Something doesn't seem right there. Historically, I've found my at home Hfe results show slightly higher than Steve's "Bare Bones" numbers. I've always just chalked this up to minor differences in test rig and if they are close i trust Steve's values.
Quote from: cnspedalbuilder on February 18, 2020, 11:10:24 AM
I did initially try the values that Steve suggested on the schematic, and those did not work. With those values, the Rangemaster sounded terrible. There was signal loss, not boost.
I think the best route is to go back to square one and check your circuit. What were the bias voltages with Steve's recommended values?
Quote from: cnspedalbuilder on February 18, 2020, 11:10:24 AM
I haven't tried plugging it in after doing all the changes I've described above, so I can certainly try that. I'll also desolder that transistor :icon_cry: and try it on the breadboard.
+1 on the breadboard. I'm an advocate of doing all testing on the breadboard before wiring everything up and making it permanent. Troubleshooting a circuit after everything is already in the board/box is always a pain in the ass.
Thank you, again, I can't say whether the Hfe value from my component tester is necessarily accurate, and it's not clear that the 79 on the packet is Steve's Hfe measurement, but he did put in resistor values suggesting that this would be a good transistor for the traditional circuit. The transistor was sitting in my garage for a year or two though, so maybe that, or the soldering, could have affected performance?
I figured that since i was using a PCB and a pre-tested transistor, this would be a cakewalk! I had no idea that the rangemaster could be so high maintenance! OK, will pull the tranny out and try it on the breadboard.
Quote from: cnspedalbuilder on February 18, 2020, 11:29:27 AM
Thank you, again, I can't say whether the Hfe value from my component tester is necessarily accurate, and it's not clear that the 79 on the packet is Steve's Hfe measurement, but he did put in resistor values suggesting that this would be a good transistor for the traditional circuit. The transistor was sitting in my garage for a year or two though, so maybe that, or the soldering, could have affected performance?
I figured that since i was using a PCB and a pre-tested transistor, this would be a cakewalk! I had no idea that the rangemaster could be so high maintenance! OK, will pull the tranny out and try it on the breadboard.
79 is the Hfe and L = Leakage.
Did you socket the transistor, or apply heat directly to the leads? If the latter, did you heat sink the transistor? It's not all that hard to damage a transistor when soldering it into a circuit.
Quote from: Big Monk on February 18, 2020, 11:38:10 AM
Did you socket the transistor, or apply heat directly to the leads? If the latter, did you heat sink the transistor? It's not all that hard to damage a transistor when soldering it into a circuit.
I was going to ask the same question. Germanium transistors are very easy to damage when soldering. I've seen some transistors installed with full lead length (insulated with tubing ) in an attempt to reduce heat conduction whilst soldering. Doing this can introduce some other fun issues that may drive you nuts.
I have some of these transistors stored away so I grabbed a couple of them and bread-boarded the circuit. No problems at all. The only adjustment done for biasing was to vary R2 (68K ). R2 was substituted using a 50K trim-pot and a 22K resistor. HFe of one transistor (AC132 ) was near 90 and the other (AC125 ) was near 60. The emitter resistor was just the standard 3k9 that is commonly used. The Boost (volume ) pot was a standard 10K.
I suspect that your transistor is very unwell.
Edit: At some point you asked if it was normal to get different resistance readings when the resistors are in circuit. The answer is yes, when measuring the in circuit resistance of a resistor your measurement will usually read lower than the value marked on the resistor. Accurate measurement is only possible when the resistor is out of circuit (or at least one end lifted out of the board ). Also never try to do a resistance measurement on a powered circuit. If you don't damage your meter by doing so, the readings that you get will be meaningless and the small amount of voltage coming from the meter will cause the circuit to behave very strangely.
Actually I did solder the transistor in directly, could that have affected the Hfe? I kept the legs on the longer side, and I'm pretty careful not to linger on those solder joints, but yes it's possible that it got hot. If the transistor is "unwell" does that mean I should just get a new one or is it still potentially useable with different resistors?
Thanks also for the tip on the meter, I did not know about that, I still have a lot to learn.
I would suggest that getting another transistor would be a good idea. There really isn't anything else that could be wrong. Try to get a tested transistor with a HFe of around 90 to 100 with low leakage - Smallbear should be able to help. I think that you got your original from there.
It is very easy to heat damage germanium transistors. Their gain and leakage properties change dramatically. See if you can find a suitable socket at Smallbear (a good quality socket ) to save having to solder the transistor. Be aware that the socket may fail after 10 to 15 years - usually corrosion or metal fatigue of the contacts.
From my quick experiments with my bread-boarded circuit and trying a few different transistors, it is worth having one of these effects....Damn, now I have yet another build to put on my list.
Quote from: cnspedalbuilder on February 19, 2020, 01:43:21 AM
...I still have a lot to learn.
Don't worry, we all do!
Quote from: Slowpoke101 on February 19, 2020, 02:05:15 AM
See if you can find a suitable socket at Small Bear (a good quality socket ) to save having to solder the transistor.
I'll second this. I usually stick a dab of hot glue on the transistors once I find the one/combo I like.
Quote from: Slowpoke101 on February 19, 2020, 02:05:15 AM
From my quick experiments with my bread-boarded circuit...
Another key point: ALWAYS breadboard first. When I was first starting out, I gave myself fits trying to troubleshoot circuits that were already hardwired/soldered into the board. A breadboard is such an essential and cheap purchase that lets you get the circuit how you want it before you fire up the soldering iron.
It also gives you a starting reference, i.e. "I had it working on the breadboard..."
I haven't yet finished breadboarding because work has been crazy but I did manage to desolder and test the transistor again. I was getting fairly different results on the component tester when I put it in, then popped it out and retested. The best and most consistent result was HfE=90, so that is good I think? Maybe the connector on the component tester is a little wonky? Or is it possble that somehow the transistor is "loose" and jiggles in and out of alignment with pressure? Maybe that is silly, I don't really know how these metal tube transistors look inside.
Quote from: cnspedalbuilder on February 21, 2020, 01:41:06 PM
I haven't yet finished breadboarding because work has been crazy but I did manage to desolder and test the transistor again. I was getting fairly different results on the component tester when I put it in, then popped it out and retested. The best and most consistent result was HfE=90, so that is good I think? Maybe the connector on the component tester is a little wonky? Or is it possble that somehow the transistor is "loose" and jiggles in and out of alignment with pressure? Maybe that is silly, I don't really know how these metal tube transistors look inside.
When you say "component tester", what do you mean exactly? You really can't just pop a Germanium transistor into say a multimeter based transistor tester because the leakage found in the Germanium unit skews the results.
If you finish breadboarding the circuit using the Small Bear transistor and the circuit sounds good, you'll know you made an error somewhere when wiring up the Aion board.
Quote from: Big Monk on February 21, 2020, 01:54:20 PM
When you say "component tester", what do you mean exactly? You really can't just pop a Germanium transistor into say a multimeter based transistor tester because the leakage found in the Germanium unit skews the results.
This is what I mean, good to know that these are likely to be off for germanium transistors!
(https://i.ebayimg.com/images/g/wpMAAOSwTEtd1gZI/s-l640.jpg)
The MK328 and similar component tester, sold on eBay and such, will read germaniums very close to what a DCA55 reads. Well worth the $15.