|
Re: Looking for TDS784A LV power supply schematic
Hi Vladimir and all
For the purpose of helping others who may have to tackle similar repairs, here is my report: It should first be noted that there are at least two types of power supplies: an older model using discrete standby circuitry, the schematic being the one mentionned by Vladimir, and a more recent one incorporating an integrated controller, the one I mentionned. Some photos of the repair can be seen here: /g/TekScopes/album?id=301696 Initial inspection revealed no voltage across the primary capacitor terminals, which required its removal to access the screw securing the bridge rectifier. It's worth mentioning that significant heat is needed to desolder the larger connections; otherwise, there's a risk of damaging the plated through-holes on the PCB. I used a soldering iron set to maximum temperature, in conjunction with a high-performance desoldering tool featuring strong suction. The Tektronix-branded PCB is of high quality and withstands heat very well. After replacing the bridge rectifier, I observed that the expected 15V rail, generated by the standby supply, was still missing. I replaced the TOP200 controller with a TOP226—a second-generation equivalent from the same manufacturer—which restored the 15V rail, but the main power supply still failed to start. I spent some time analyzing the protection circuitry. Disconnecting the various protection subsystems one by one did not resolve the issue. The supply remained stuck in a "hiccup" mode—operating for approximately 10 milliseconds once per second. This short active window allowed for probing and measuring voltages on the power rails. To better monitor system status, I added LED indicators to the control circuitry. The first LED confirmed presence of the 15V standby voltage. A second LED, triggered via a transistor, illuminated when the voltage on pin 10 of the 3525 PWM controller (available at the cathode of CR9 near the top of the A18 board) was high—indicating that either the overvoltage or overcurrent protection had activated, or that the power supply was shut down. I then cut away the portion of the PCB that had been completely burned beneath capacitors C47 and C83. Unfortunately, several surrounding traces were also affected by the damage and had to be removed and re-routed using insulated Teflon wire. Once repairs were completed, the LED indicators confirmed that the unit was still operating in hiccup mode. Nonetheless, I was able to take voltage readings during the brief on-times. These preliminary measurements were mostly within expected ranges. Next, I connected resistive loads to the dual-row headers labeled P5 and P6: 12-ohm, 2-watt resistors for the 5V outputs, 100 ohms for the 15V lines, and 330 ohms for the 24V rail. With these loads in place, the power supply began operating continuously, enabling accurate measurement of output voltages. All outputs were within spec, except the 24V line, which measured 32V. Since the 24V output is unregulated and rated for a nominal 2.5A, I connected an ammeter and a 10-ohm load to draw current. Under load, the voltage dropped to 22.5V, which I deemed acceptable. To mechanically reinforce the repair, I used silicone adhesive to secure the replacement capacitors to one of the power supply's metal supports. After reinstalling the unit, I was pleased to find that the device powered on successfully and passed all self-tests. |
Jean-Paul
Jose Luu