TI unveils complete 800 VDC power architecture for future generation AI data centers with NVIDIA

Rhea-AI Summary

Texas Instruments (NASDAQ: TXN) unveiled a complete 800 VDC power architecture for next-generation AI data centers in collaboration with NVIDIA, demonstrated at NVIDIA GTC March 16-19, 2026. The design uses only two conversion stages (800V→6V→1V) and highlights a 800V→6V converter and >2000W/in3 power density.

TI also showed a 30kW 800V AC/DC PSU, 800V capacitor bank units with 40W/in3 using EDLC cells, and an 800V hot-swap controller supporting scalable high-voltage racks.

Key Figures

Bus voltage: 800 VDC Conversion stages: 2 stages Intermediate bus voltage: 6V +5 more

8 metrics

Bus voltage 800 VDC Power architecture for next-generation AI data centers

Conversion stages 2 stages From 800V to GPU core power in TI architecture

Intermediate bus voltage 6V 800V to 6V isolated bus converter before GPU core stage

GPU core voltage <1V Output of multiphase buck converter for GPU cores

Peak efficiency 97.6% 800V to 6V DC/DC bus converter

Power density (bus) >2000 W/in3 800V to 6V DC/DC bus converter for compute trays

PSU power rating 30 kW 800V high power density AC/DC PSU for AI servers

Capacitor bank density 40 W/in3 800V capacitor bank units using EDLC super capacitors

Market Reality Check

Price: $194.45 Vol: Volume 4,735,439 is below...

normal vol

$194.45 Last Close

Volume Volume 4,735,439 is below the 20-day average of 6,408,098 (relative volume 0.74). normal

Technical Price 194.13 is trading above the 200-day MA at 190.28, indicating a pre-news uptrend bias.

Peers on Argus

TXN was up 2.13% while key peers were mixed: ARM up, but MU, QCOM, ADI and INTC ...

TXN was up 2.13% while key peers were mixed: ARM up, but MU, QCOM, ADI and INTC down. Moves appear stock-specific rather than a broad semiconductor rotation.

Previous AI Reports

4 past events · Latest: Mar 10 (Positive)

Same Type Pattern 4 events

Date	Event	Sentiment	Move	Catalyst
Mar 10	Edge AI MCUs launch	Positive	+0.6%	Introduced MCUs with TinyEngine NPU and expanded edge AI software ecosystem.
Mar 05	NVIDIA AI robotics collab	Positive	-2.2%	NVIDIA collaboration to enable safe humanoid robots with TI sensing and control.
Jan 06	Auto edge AI chips	Positive	+0.9%	Launched edge AI radar and audio processors for enhanced in‑cabin automotive features.
Nov 11	Real-time AI MCUs	Positive	-2.0%	New real-time MCUs with edge AI accelerators targeting efficiency and safety gains.

Pattern Detected

AI-related announcements have produced mixed reactions, with similar AI news yielding both modest gains and declines.

Recent Company History

Over the past year, Texas Instruments has steadily expanded its AI footprint across edge, automotive and industrial markets. Prior AI-tagged news included new MCUs with TinyEngine NPU for edge AI, a collaboration with NVIDIA on humanoid robotics, and automotive edge-AI radar and MCUs. Those events saw small positive and negative moves, suggesting the market has treated AI announcements as incremental rather than transformative. Today’s 800 VDC AI data center architecture extends this multi-domain AI strategy.

Historical Comparison

-0.7% avg move · Past AI-tagged announcements moved TXN by an average of -0.67%, showing generally muted, mixed react...

AI

-0.7%

Average Historical Move AI

Past AI-tagged announcements moved TXN by an average of -0.67%, showing generally muted, mixed reactions to similar AI product and collaboration news.

AI news has progressed from edge and automotive MCUs and sensors to collaborations with NVIDIA and now to an 800 VDC power architecture for AI data centers, broadening TI’s role across the AI compute stack.

Market Pulse Summary

This announcement highlights TI’s push into AI data center infrastructure with an 800 VDC architectu...

Analysis

This announcement highlights TI’s push into AI data center infrastructure with an 800 VDC architecture, featuring two conversion stages, up to 97.6% efficiency and power densities above 2000 W/in3. It extends a pattern of AI-focused launches across edge, automotive and industrial markets. Investors may track adoption of these reference designs with NVIDIA, the pace of AI data center buildouts, and how TI integrates this platform with its broader analog and embedded portfolio.

Key Terms

multiphase buck converter, ac/dc psu, edlc, super capacitor, +1 more

5 terms

multiphase buck converter technical

"6V to <1V multiphase buck converter: High-current solution for advanced GPU cores"

A multiphase buck converter is an electrical component that turns a higher direct current (DC) voltage into a lower DC voltage using several synchronized switching channels working in parallel to share the load. It reduces voltage ripple, spreads heat, and raises efficiency compared with a single-channel regulator, so devices run cooler and use less energy. For investors, that means products can deliver higher performance, longer battery life and lower cooling or component costs—factors that influence competitiveness and margins—much like several people taking turns carrying a heavy load to keep the pace steady and avoid overload.

ac/dc psu technical

"a 30kW 800V high power density AC/DC PSU for AI servers"

An ac/dc PSU is an electrical device that converts the alternating current (AC) from a wall outlet into the direct current (DC) that electronics need to run, like a phone charger or laptop adapter on a larger or industrial scale. Investors care because these units are critical components in many products and supply chains; changes in demand, cost, or availability can affect manufacturers’ production, margins, and revenue.

edlc technical

"800V capacitor bank units (CBU) with 40W/in3 power density using EDLC super capacitor"

An EDLC (electric double-layer capacitor) is a type of energy-storage device that holds charge on surfaces rather than inside chemical reactions, so it charges and discharges very quickly and lasts through many cycles. For investors, EDLCs matter because they can enable faster charging, longer equipment life, and different product designs in industries like renewable power, transportation, and electronics, which can shift revenue opportunities and competitive advantage without replacing batteries entirely.

super capacitor technical

"800V capacitor bank units (CBU) with 40W/in3 power density using EDLC super capacitor"

A super capacitor is an electrical energy storage device that charges and discharges much faster than a battery, like a high-capacity spring that can be wound and released quickly. Investors care because it enables rapid power delivery and long service life for products from electric vehicles to grid equipment, potentially lowering maintenance and improving performance; companies that master the technology can gain cost or efficiency advantages in power-dependent markets.

hot-swap controller technical

"800V hot-swap controller: Scalable hot-swap solution for input power protection"

A hot-swap controller is an electronic component that safely manages the connection and removal of power-hungry modules (like drives or power supplies) while a system stays running, preventing sparks, surges, or data loss. For investors, it matters because reliable hot-swap functionality reduces downtime, maintenance costs, and risk of equipment damage—factors that affect product quality, service continuity and the long-term operating costs of companies selling or using server, telecom, or storage hardware.

AI-generated analysis. Not financial advice.

TI's complete power solution includes multiple breakthrough reference designs with industry-leading specification

News highlights:

• TI has developed a complete 800 VDC power solution for future generation AI data centers with NVIDIA
• As part of this collaboration, TI is demonstrating a power architecture requiring only two conversion stages from 800V to processor power
• TI will demonstrate its 800 VDC power solution at NVIDIA GTC 2026

DALLAS, March 16, 2026 /PRNewswire/ -- Texas Instruments (NASDAQ: TXN) today unveiled a complete 800V direct current (DC) power architecture for next-generation AI data centers built with the NVIDIA 800 VDC reference design. The solution will be showcased at NVIDIA GTC, March 16-19, 2026, at NVIDIA's power architecture display and TI's booth 169, demonstrating how TI's analog and embedded processing technology supports NVIDIA's vision for advancing high-voltage systems in AI data centers.

"The exponential growth of AI computing demands a fundamental rethinking of how we deliver power in data centers," said Kannan Soundarapandian, vice president and general manager of high-voltage power at TI. "Our advanced 800 VDC architecture represents a critical breakthrough that enables data center operators to meet today's power challenges while preparing for tomorrow's AI workloads. By collaborating with NVIDIA, we're helping accelerate the deployment of AI infrastructure that can scale efficiently and reliably."

Addressing critical power challenges in AI infrastructure
As AI workloads continue to drive unprecedented power requirements in data centers, traditional power distribution architectures are reaching their limits. TI's 800 VDC architecture addresses these challenges by maximizing conversion efficiency and power density across the entire power path, simplifying the power architecture and enabling more scalable and reliable AI data center operations.

TI's breakthrough approach requires only two conversion stages from 800V to GPU core power: compact 800V to 6V isolated bus converter with higher peak efficiency, followed by a 6V to <1V multiphase buck solution with high current density generation-over-generation. This streamlined architecture supports the NVIDIA reference design.

Complete power solution 
TI's comprehensive 800 VDC power architecture solution shown at NVIDIA GTC includes multiple breakthrough reference designs with industry-leading specification:

• 800V hot-swap controller: Scalable hot-swap solution for input power protection for 800V rails.
• 800V to 6V DC/DC bus converter: High density solution featuring integrated GaN power stages, delivering 97.6% peak efficiency with >2000W/in³ power density for compute tray applications.
• 6V to <1V multiphase buck converter: High-current solution for advanced GPU cores, with higher power density vs 12V designs and featuring dual-phase power stages.

In addition to these designs, TI will showcase a 30kW 800V high power density AC/DC PSU for AI servers and 800V capacitor bank units (CBU) with 40W/in³ power density using EDLC super capacitor cells. TI will also be featuring an 800V to 12V DC/DC bus converter for compute tray power conversion.

Accelerating development of future generation AI data centers
As the data center industry experiences growth driven by AI and edge computing, TI's analog and embedded processing technologies deliver what's needed to efficiently power and manage advanced AI workloads. From solid-state transformers to sidecar and server IT racks, and cooling distribution units, TI's scalable semiconductor solutions help pave the way for future AI infrastructure deployments.

About Texas Instruments 
Texas Instruments Incorporated (Nasdaq: TXN) is a global semiconductor company that designs, manufactures and sells analog and embedded processing chips for markets such as industrial, automotive, data center, personal electronics and communications equipment. At our core, we have a passion to create a better world by making electronics more affordable through semiconductors. This passion is alive today as each generation of innovation builds upon the last to make our technology more reliable, more affordable and lower power, making it possible for semiconductors to go into electronics everywhere. Learn more at ti.com.

View original content to download multimedia:https://www.prnewswire.com/news-releases/ti-unveils-complete-800-vdc-power-architecture-for-future-generation-ai-data-centers-with-nvidia-302714861.html

SOURCE Texas Instruments

FAQ

What is Texas Instruments' (TXN) 800 VDC architecture announced March 16, 2026?

TXN announced a complete 800 VDC power architecture for AI data centers, using two conversion stages from 800V to GPU core power. According to the company, it supports an 800V→6V isolated bus and a 6V→<1V multiphase buck to maximize efficiency and power density.

How efficient is the TI 800V to 6V DC/DC converter in the TXN announcement?

The 800V→6V DC/DC bus converter achieves a reported 97.6% peak efficiency. According to the company, this high efficiency pairs with >2000W/in3 power density for compute tray applications, designed to reduce conversion losses in AI server power paths.

What conversion stages does TXN's NVIDIA collaboration recommend for AI data centers?

TI's recommended architecture uses only two conversion stages: 800V to 6V isolated bus, then 6V to <1V multiphase buck for GPU cores. According to the company, this streamlined path simplifies design while improving generation-over-generation current density.

What hardware did Texas Instruments (TXN) showcase at NVIDIA GTC 2026?

TI showcased multiple reference designs including an 800V hot-swap controller, a 97.6% efficient 800V→6V converter, a 6V→<1V multiphase buck, a 30kW 800V AC/DC PSU, and 800V capacitor bank units. According to the company, these target high-density AI compute trays and racks.

How does TXN claim the 800 VDC approach affects power density for AI servers?

TI claims the 800 VDC approach increases power density versus traditional 12V designs, citing >2000W/in3 for the 800V→6V converter and 40W/in3 for 800V capacitor bank units. According to the company, these metrics enable more compact, higher-power compute trays.

Will TXN's 800 VDC architecture work with NVIDIA's reference design and AI GPUs?

Yes — TI's architecture is presented to support NVIDIA's 800 VDC reference design and advanced GPU power needs. According to the company, the two-stage 800V→6V→<1V path is built specifically to align with processor power requirements in next-generation AI data centers.