When you assemble your next board, is it possible that you don't have the fasteners so close to each other as they are now?
You are mounting the board with those two fasteners between the Q1, Q2 and IC1, IC2. When you tighten the nuts, it could be pulling them together.
Full B+ voltage is on IC1 metal tab. Regardless of insulation to not, it's still not good if IC1 and Q1 are touching, don't you think? Free insurance if they're not touching.
You are mounting the board with those two fasteners between the Q1, Q2 and IC1, IC2. When you tighten the nuts, it could be pulling them together.
Full B+ voltage is on IC1 metal tab. Regardless of insulation to not, it's still not good if IC1 and Q1 are touching, don't you think? Free insurance if they're not touching.
When I set up the new board I will run one of the sets of bolts in the opposite direction to avoid them being so close. Thanks for the feedback.
Hoping things so more smoothly this time around. Looking forward to more music and less smoke
Received and built another board today, thanks George! Had no issues this time around. The checkout process went smoothly and now im here listening to some awesome music through the amp. I had one last question for everyone, Im running GL 300b tubes at 400v. Any recommendations on the plate current that will sound best? Feeding it into 5k OPT. Thanks!
Hey Bryce, glad you persevered and got it sorted out. BTW, I saw the chassis you've built for it in another thread; good job! 
I can't speak for 300Bs, but I've been running a pair of Chinese 2A3s (I call them 2A300Bs) at about 70 mA now for six months with no issues. Just for the record, I usually run genuine 2A3s at about 49 mA and 45s at ~30.
I can't speak for 300Bs, but I've been running a pair of Chinese 2A3s (I call them 2A300Bs) at about 70 mA now for six months with no issues. Just for the record, I usually run genuine 2A3s at about 49 mA and 45s at ~30.
Hey Bryce, glad you persevered and got it sorted out. BTW, I saw the chassis you've built for it in another thread; good job!
I can't speak for 300Bs, but I've been running a pair of Chinese 2A3s (I call them 2A300Bs) at about 70 mA now for six months with no issues. Just for the record, I usually run genuine 2A3s at about 49 mA and 45s at ~30.
Thanks, it was a fun project and I was extra determined to get it working after having issues with the first build. Everyone on the forum is super helpful and helped give me the confidence to get it done
I strongly suspect that the flux in the solder used on the first board absorbed some moisture from the wipe down. This could have caused some of the smoking experiences during initial checkout.
I have experimented on the board you sent back, even to the point of applying 800 volts across some of the places that arcing originally occurred, with no fireworks. The white crusty looking residue that was visible in your pictures is no longer present, and I believe that was the real issue.
To test this theory I ordered some of the exact same solder and I will test it once it arrives, but that won't be until the middle of next week.
That would make sense, the board also sat in the garage for about two months after being built while I waited for a transformer. It would have had plenty of time and opportunity to absorb moisture. So the flux residue absorbed moisture which caused the shorts that caused the resistor to blow.
That's the most logical explanation that I can come up with at the moment. I will know more if I can duplicate the fireworks on a new board once I have the solder in hand. The literature states that the flux is water soluble. That might not make much difference on a 5 volt Arduino board, but may not be ideal for something with 350 to 450 volts present.
I didn't even know it was possible for flux to become conductive but after the fireworks show I spent a few hours reading online and while not common, it is possible under the right(or wrong) conditions. Lesson learned and Im grateful I didn't wreck any of the iron I had bought for the amp. Thanks again George for the awesome customer service and getting me a board shipped out same day.
When Freon was banned from the (Motorola) factory where I worked in the 70's they tried some water based stuff that made the factory smell like an orange grove. We had all sorts of strange issues with products that used it, and the highest voltages seen in the products was 15 volts. The machine operators couldn't stand the smell, so we went to a no-clean flux. Then SMD came along in the 80's and everything changed.
how was the Freon used to clean the flux? Was it just poured onto the board and scrubbed?
We had fancy machines called vapor phase degreasers. I was not responsible for these machines, but I did watch several teardowns, and used them for everything from DIY PC boards to dirty spark plugs. Simple explanation:
Freon in vapor form is heavier than air. It collects in the bottom of a closed box at a pretty high density, and is strong enough to remove flux. The boards enter a tall metal box on a webbed steel belt through an opening on one side and pass through and out an opening on the other side. There is liquid Freon in the bottom of the box which is constantly heated to a boil such that Freon vapors are rising up through the box passing around the boards. There are focused jet streams for known problem areas on the boards. In the top of the box there are refrigeration coils that turn the vaporized Freon back to a liquid and return it to the bottom of the box for filtration and reuse.
Freon in vapor form is heavier than air. It collects in the bottom of a closed box at a pretty high density, and is strong enough to remove flux. The boards enter a tall metal box on a webbed steel belt through an opening on one side and pass through and out an opening on the other side. There is liquid Freon in the bottom of the box which is constantly heated to a boil such that Freon vapors are rising up through the box passing around the boards. There are focused jet streams for known problem areas on the boards. In the top of the box there are refrigeration coils that turn the vaporized Freon back to a liquid and return it to the bottom of the box for filtration and reuse.
Maybe, but all of this happened in the late 70's early 80's. You could get Freon R12 for your car AC, and Freon based cleaning products at the auto parts stores.
For unknown reasons not stated, Digikey committed more fumbles with my order than happened in the Super bowl, so I won't have my parts and the solder until maybe Friday or Saturday, and testing may not occur until next Tuesday.
I got the order in early enough on Friday afternoon that it should have shipped that day. All items were in stock, and I got an order confirmation email stating this. No shipment confirmation appeared, and the web site only said "order submitted" on Saturday and Sunday.
Sunday afternoon I get an email that states "We have experienced an inventory discrepancy on your Digi Key Sales Order XXXX. The following has been placed on backorder and will ship once they are available." Estimated delivery date in in July (mosfets for a power supply). 20 minutes later I get another email with the invoice and a tracking number Sunday around 5 PM.
Monday I get another email, "Due to an internal delay, your order placed with Digi-Key was unable to ship as expected." again with no information or explanation.
This morning the tracking number is finally active, but the package is not yet in the hands of the USPS.
Seems like the two hands shook, and settled their differences. The parts showed up this morning with a totally different tracking number that I never got.
There are still 4 fets missing from the order....I guess that I won't get to blow them up. They were for power supply experiments at 650 volts where things often go wrong, really wrong. Curve tracing a BIG tube well beyond it's 40 watt ratings needs a power supply with quick and accurate current limiting. So far the limiter is blown parts!
There are still 4 fets missing from the order....I guess that I won't get to blow them up. They were for power supply experiments at 650 volts where things often go wrong, really wrong. Curve tracing a BIG tube well beyond it's 40 watt ratings needs a power supply with quick and accurate current limiting. So far the limiter is blown parts!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.