I found a real 75W bulb and put in the mains tester. I brought it back up and got a light again, but had a screwdriver in the offset and turn it quickly and the light went out. When I turn it back to the left the light begins to come on again.
There is a 0mV hole in the board with one side going through the resistor to one leg of the R100. The other hole appears isolated. This is DC offset? (Measure the non-isolated to ground?)
There is a 0mV hole in the board with one side going through the resistor to one leg of the R100. The other hole appears isolated. This is DC offset? (Measure the non-isolated to ground?)
Okay,, if that 0mV hole to ground is DC offset, it's working, but when I get close to zero the light come on full and the offset goes nonsensical. What I think guys is this thing is working as it should be, and I just need to take the damned mains bulb out of the circuit, but before I do, I want some input from Jeff or someone who knows more than I. Also, where is the best place to verify bias voltage.
Thanks guys, I appreciate ya all!!
JT
Thanks guys, I appreciate ya all!!
JT
New fangled bulbs are useless as a bulb tester -- you need the resistance of the old-school filament.
The 0mV holes are so that you can solder in a bent resistor lead to have a handy attachment point for a probe to measure offset. (It sounds like you figured that out, so I'm just confirming.)
The bias is fixed on the J2; the trimmer just controls the offset. So I wouldn't think it would have any effect on the bulb. Hopefully someone more knowledgeable will reply....
The 0mV holes are so that you can solder in a bent resistor lead to have a handy attachment point for a probe to measure offset. (It sounds like you figured that out, so I'm just confirming.)
The bias is fixed on the J2; the trimmer just controls the offset. So I wouldn't think it would have any effect on the bulb. Hopefully someone more knowledgeable will reply....
Roger on the bulb, I figured as much. Also, I did solder in a probe too. As for the DC offset, what I see is when the offset is really high, the light is off or dim, but as I sneak-up on zero the light begins to glow and just about the time I would reach zero the light is full bright and then the offset goes bonkers.
Still looking for some thoughts on this one people, I don't want to pull that bulb until I am relitively sure she won't smoke. I now have a 72W old school bulb in there. I did notice when I was building the F5T and had a 100W bulb in there, that the bulb would start to come on when I got close to full bias. I think I also witnessed the random fluctuations in the DC offset when this happened as well. I'm not positive, though, so I will give you guys a chance to set me straight...
Please
JT
Still looking for some thoughts on this one people, I don't want to pull that bulb until I am relitively sure she won't smoke. I now have a 72W old school bulb in there. I did notice when I was building the F5T and had a 100W bulb in there, that the bulb would start to come on when I got close to full bias. I think I also witnessed the random fluctuations in the DC offset when this happened as well. I'm not positive, though, so I will give you guys a chance to set me straight...
Please
JT
About where is the pot set when the offset reaches 0?
You should be a full bias on the J2 at all times. So the lightbulb should be glowing a bit, just not "on".
It's possible there's some interaction between the bulb and the amp; I use a variac instead so I'm not familiar with the bulb tester.
You should be a full bias on the J2 at all times. So the lightbulb should be glowing a bit, just not "on".
It's possible there's some interaction between the bulb and the amp; I use a variac instead so I'm not familiar with the bulb tester.
Okay, I got it to about 1mV, twitch little sucker the light was on, and bright, but not full bright I don't think. I would like to check to see what the bias is, I know you can't change it, but just to make sure it in the ballpark of where it should be. DO I do that across R5-9?
Also, I noticed a call out on your print off the base of the 170s of 9.5mA, but also thought I saw a correction to 8.5mV. If ya cam Jeff, give me a couple test points to do some sanity checks.
Also, I noticed a call out on your print off the base of the 170s of 9.5mA, but also thought I saw a correction to 8.5mV. If ya cam Jeff, give me a couple test points to do some sanity checks.
The sweet-spot on the LTP current is pretty shallow, so anything between 6 and 12mA will do (I think I remember Nelson saying the factory ones used between 8 and 12mA). I calculated 8.5mA as the "sweetest" spot.
Measure the voltage across R8 to check the bias. Apply Ohm's law to whatever resistance you used and you should get 1.3 - 1.4A.
Measure the voltage across R8 to check the bias. Apply Ohm's law to whatever resistance you used and you should get 1.3 - 1.4A.
I'm getting about half that with the light on. I just checked the rails and with that light on they are only sitting at about 5v. The bias is 75mA, but with the rails at 5v I see why.
I think the light may not be letting it pull the current it needs, but it could also be keeping the smoke in, I shall wait and hope for enlightenment.
I think the light may not be letting it pull the current it needs, but it could also be keeping the smoke in, I shall wait and hope for enlightenment.
Last edited:
I did manage to find a 150W bulb for the tester and it was on, but very dim, so....
I decided to grow a pair and go to mains.... I need to get my laser temp out those suckers get hot quick. I also found that I am seeing about 511mV across the .5R resistor, so it looks to be a little low no?
EDIT: Didn't have it on long... about 5mins or so and the SEP100 were running about 26c
I'm still a little gun-shy, but at least it didn't smoke straight-away. Resistor values are per your recommendations, not sure why I am almost .5 short on the bias. 512mV/.5R~1A
Any thoughts Jeff?
JT
Last edited:
26ºC is stone cold. Mine run at something like 60ºC.
The JFET has a positive temp coefficient so it will conduct more as it warms up. The opto-coupler counteracts that to keep it from running away, but it won't be an exact correction. I've never measured it to see if it under-corrects or over-corrects, but if it under-corrects then you'll likely end up at 1.3 or 1.4A when it warms up.
The JFET has a positive temp coefficient so it will conduct more as it warms up. The opto-coupler counteracts that to keep it from running away, but it won't be an exact correction. I've never measured it to see if it under-corrects or over-corrects, but if it under-corrects then you'll likely end up at 1.3 or 1.4A when it warms up.
Sounds good Jeff, I'm working on the sesond channel now. I will complete that hook it up solo and check the reading to see if they are in the ball park, then reconnect the other channel.
It's hard to be a FAB when you have limited funds and unobtainium parts! I came home from work today dropped in the 150W bulb had a look at that and just thought well, if not now then when? After all, I got to get this one done, so I can do the AN. lol
Then depending on which one I like better in the system, one of the projects will be turned into monos and take their place in the line-up.
Thanks for all your help Jeff.
Cheers!
JT
Last edited:
Changing sink from vertical to Horiz
Hint: For those of you yet to build. I used the standard sink from the DIY store. I wanted to mount horiz, so I called a local metal shop and had them cut a piece of 3/8 angle iron to the appropriate size. You don't need to go full length of the sinks, or PCB for that matter. Next, I took it down to a friends machine shop and cleaned up two of the four surfaces. The one that touches the sink and the one that the chip sinks too. I get them all nice and true and flat, but still had some tool marks, I could have let it go there, but we are DIY guys, so I got some 800 and a pad sander... ya need to do this wet, or it will load you paper. After some work they were looking nice and I finished it off with 1000. Truth be told 800 would have been more than enough, but again... diy.
Okay, so I'm all set to drill some holes for mounting? No! Remember the template that Jeff gives you works for a solid one piece sink. Since you are using two bolted together this will NOT work. There is always going to be a bit of misalignment in both height and width. There's an easy fix to get your angle iron to sit flush. Since there is no need to mount the resistors to the sink( See Jeff's earlier post) this give you a lot of leeway to move the PCB forward or back. Simply cut your angle iron in half and remove another 1/8 to 1/4 from each side where your sink to sink seem will be bolted. (the 1/8 gives you the ability to stay away from the seem in your sink) Now, you can bolt your angle to you sinks taking care to place the PCB or template to make sure you are good for your case and also for the placement of the JFETs. Once you get them bolted on (I used used old school silicon white between the angle and sink) you may now lay your template or PCB on the angle and mark it. I found this to be a little off when using a center punch and drill. So when I did the second, I simply clamped the PCB on the angle and measured to make sure the leads resistor leads would not contact the angle. I also measured off the rear of the case about 7/8ths. The measurement to keep the leads from making contact with the angle iron/aluminum. The space will be about 3/4. IN other words, the pcb will hang off the angle about 3/4 of an inch into the air.
With the PCB firmly clamped to the top of the angle, I used the PCB as a guide to make sure the holes were where the should be. I drilled one hole in the rear and then lightly bolted it, so it could not move and drilled the front hole, pinned that and drill both hols for the JFET. Next, I pulled the PCB off and drilled the holes a bit larger to give me a slight bit of adjustment in case they were slightly off.
After that, I vacuumed up all the chips and transferred the PCB to the bottom of the angle where it will reside. The holes were perfect... this process worked much better than the first rail.
Another hint: Since we are not sinking the resisters, to the main sink, I used snap-ons, You can use the one hole on the template/PCB made for mounting it to the sink, and on the other end, you use the hole for the resistor. Works slick as hell. Wish I had seen all this on the first, one... Hope this may help someone down the line.
If anyone is going this route and needs images, please let me know and I will take some pics to illustrate.
EDIT: Don't forget to chamfer your holes to remove any metal burs that may inhibit your sink angle to sink and JFET to angle.
JT
Hint: For those of you yet to build. I used the standard sink from the DIY store. I wanted to mount horiz, so I called a local metal shop and had them cut a piece of 3/8 angle iron to the appropriate size. You don't need to go full length of the sinks, or PCB for that matter. Next, I took it down to a friends machine shop and cleaned up two of the four surfaces. The one that touches the sink and the one that the chip sinks too. I get them all nice and true and flat, but still had some tool marks, I could have let it go there, but we are DIY guys, so I got some 800 and a pad sander... ya need to do this wet, or it will load you paper. After some work they were looking nice and I finished it off with 1000. Truth be told 800 would have been more than enough, but again... diy.
Okay, so I'm all set to drill some holes for mounting? No! Remember the template that Jeff gives you works for a solid one piece sink. Since you are using two bolted together this will NOT work. There is always going to be a bit of misalignment in both height and width. There's an easy fix to get your angle iron to sit flush. Since there is no need to mount the resistors to the sink( See Jeff's earlier post) this give you a lot of leeway to move the PCB forward or back. Simply cut your angle iron in half and remove another 1/8 to 1/4 from each side where your sink to sink seem will be bolted. (the 1/8 gives you the ability to stay away from the seem in your sink) Now, you can bolt your angle to you sinks taking care to place the PCB or template to make sure you are good for your case and also for the placement of the JFETs. Once you get them bolted on (I used used old school silicon white between the angle and sink) you may now lay your template or PCB on the angle and mark it. I found this to be a little off when using a center punch and drill. So when I did the second, I simply clamped the PCB on the angle and measured to make sure the leads resistor leads would not contact the angle. I also measured off the rear of the case about 7/8ths. The measurement to keep the leads from making contact with the angle iron/aluminum. The space will be about 3/4. IN other words, the pcb will hang off the angle about 3/4 of an inch into the air.
With the PCB firmly clamped to the top of the angle, I used the PCB as a guide to make sure the holes were where the should be. I drilled one hole in the rear and then lightly bolted it, so it could not move and drilled the front hole, pinned that and drill both hols for the JFET. Next, I pulled the PCB off and drilled the holes a bit larger to give me a slight bit of adjustment in case they were slightly off.
After that, I vacuumed up all the chips and transferred the PCB to the bottom of the angle where it will reside. The holes were perfect... this process worked much better than the first rail.
Another hint: Since we are not sinking the resisters, to the main sink, I used snap-ons, You can use the one hole on the template/PCB made for mounting it to the sink, and on the other end, you use the hole for the resistor. Works slick as hell. Wish I had seen all this on the first, one... Hope this may help someone down the line.
If anyone is going this route and needs images, please let me know and I will take some pics to illustrate.
EDIT: Don't forget to chamfer your holes to remove any metal burs that may inhibit your sink angle to sink and JFET to angle.
JT
Last edited:
@Jeff,
I got the second channel up today, no problems at all. The DC offset was very easy to set on this side, not twitchy at all. That makes me think there is something going on with the first channel. How was yours when you set the offset, the first channel was all over the place and wondered around a bit. THe second was rock solid and easy to set.
One thing I did notice is the pot itself on the first board feel loose, but on the second it has more friction and a smoother feel. Perhaps I have an issue with that pot. I do have a couple extra. (They are Bournes)
THoughts?
JT
I got the second channel up today, no problems at all. The DC offset was very easy to set on this side, not twitchy at all. That makes me think there is something going on with the first channel. How was yours when you set the offset, the first channel was all over the place and wondered around a bit. THe second was rock solid and easy to set.
One thing I did notice is the pot itself on the first board feel loose, but on the second it has more friction and a smoother feel. Perhaps I have an issue with that pot. I do have a couple extra. (They are Bournes)
THoughts?
JT
Last edited:
Will do on the pot.
Also, those big honking pieces of Aluminum angle have done some crazy things to the temp... I looked at your numbers and the right channel has been running a couple hours now and is at 36.4C hehe The resistors which just have the snap-on are running about the same as the JFETs, so I would advise anyone who does this to at least get some snap-on sinks, I think those things would be to hot to run naked.
Also, those big honking pieces of Aluminum angle have done some crazy things to the temp... I looked at your numbers and the right channel has been running a couple hours now and is at 36.4C hehe The resistors which just have the snap-on are running about the same as the JFETs, so I would advise anyone who does this to at least get some snap-on sinks, I think those things would be to hot to run naked.